development of a new mix design method and specification
TRANSCRIPT
University of Texas at El PasoDigitalCommons@UTEP
Open Access Theses & Dissertations
2011-01-01
Development Of A New Mix Design Method AndSpecification Requirements For Asphalt TreatedBasesHector Antonio HernandezUniversity of Texas at El Paso, [email protected]
Follow this and additional works at: https://digitalcommons.utep.edu/open_etdPart of the Civil Engineering Commons
This is brought to you for free and open access by DigitalCommons@UTEP. It has been accepted for inclusion in Open Access Theses & Dissertationsby an authorized administrator of DigitalCommons@UTEP. For more information, please contact [email protected].
Recommended CitationHernandez, Hector Antonio, "Development Of A New Mix Design Method And Specification Requirements For Asphalt TreatedBases" (2011). Open Access Theses & Dissertations. 2501.https://digitalcommons.utep.edu/open_etd/2501
DEVELOPMENT OF A NEW MIX DESIGN METHOD AND SPECIFICATION
REQUIREMENTS FOR ASPHALT TREATED BASES
HECTOR A. HERNANDEZ
Department of Civil Engineering
APPROVED:
Soheil Nazarian, Ph.D., Chair
Vivek Tandon, Ph.D.
Peter Golding, Ph.D.
Patricia D. Witherspoon, Ph.D. Dean of the Graduate School
Copyright ©
by
Hector A. Hernandez
2011
Dedication
This thesis is dedicated to the most beloved people in my life who supported me and gave strength throughout my studies.
My Mother Guillermina Sandoval
My Father Humberto Antonio Hernandez
My Brothers Humberto and Jose Luis Hernandez
And to my Wife Paulina Espinosa de los Monteros
DEVELOPMENT OF A NEW MIX DESIGN METHOD AND SPECIFICATION
REQUIREMENTS FOR ASPHALT TREATED BASES
by
HECTOR A. HERNANDEZ, BSCE
THESIS
Presented to the Faculty of the Graduate School of
The University of Texas at El Paso
in Partial Fulfillment
of the Requirements
for the Degree of
MASTER OF SCIENCE
Department of Civil Engineering
THE UNIVERSITY OF TEXAS AT EL PASO
May 2011
v
Acknowledgements
I am grateful to all of the people who helped and guided me throughout the years.
Without their support this project could not have been completed.
My mentor, Dr. Soheil Nazarian, provided continuous guidance and support throughout
my career at UTEP. I thank him for his invaluable counseling.
I also want to thank my committee members, Dr. Vivek Tandon and Dr. Peter Golding.
Each has provided me with great help and assistance along my studies at UTEP.
Special thanks to Mr. Jose Garibay, without his help and guidance this study could not be
done.
I would like to thank Dr. Manuel Celaya and Dr. Imad Abdallah for his continuous help
and guidance throughout this study. I thank them for his invaluable friendship.
I would like to acknowledge the funders of my study, the Texas Department of
Transportation (TxDOT especially to Jimmy Si, Project Director, for his outstanding support and
guidance on this project.
I would like to thank my family that always encouraged me and support me throughout
all my studies. I would also like to thank my friends and co-workers, Jose Luis Hernandez,
Carlos Manzanera, Alejandro Castillo, Jose Moran, Leslie Ponce, Eric Navarro, Carlos Solis, and
Tania Ramirez.
vi
Abstract
Asphalt stabilized bases in Texas are usually designed and constructed as per Item 292,
“Asphalt Treatment (Plant Mixed),” of the 2004 Standard Specification book. This specification
is a hybrid of base and hot mix asphalt concrete procedures and requirements, which are
sometimes incompatible. In addition, this Item uses a specific Texas Gyratory compactor that is
not readily available to all districts. Some districts have started using test method Tex-204-F,
Part III, ‘Mix Design for Large Stone Mixtures Using the Superpave Gyratory Compactor.”
However, this procedure was originally developed to design Type A and Type B hot mix at 96%
density and produce a 6 in. by 4.5 in. specimen. Under Item 292, the unconfined compressive
strength of the mix (as per Tex-126-E) is used to assess the quality of the mix. Specimens
prepared under Tex-204-F are not the appropriate size for this type of testing. As such, the
quality of the mix is assessed with the indirect tensile strength. A new mix design procedure is
needed for this type of material that can use standard equipment such as the Superpave Gyratory
Compactor (SGC) to mold the specimens for mix design.
In order to achieve the objective of this project, current TxDOT procedures such as Tex-
126-E and Tex-204-F were evaluated and modified in order to propose new Tex-126-H and Tex-
204-H specifications. A comprehensive parametric study comparing the results of the two
proposed specifications was performed. The impact of the number of gyrations, curing
temperature, binder grade, and asphalt content variation were evaluated using prepared
laboratory specimens. Parameters including density, unconfined compressive strength, indirect
tensile strength, and modulus using the existing and proposed specifications were compared.
Based on these studies, a new method for determining the Optimum Asphalt Content (OAC) for
ATBs was developed. The recommendations were then evaluated at six actual construction
projects for reasonableness.
The most practical set-up for laboratory tests was achieved using Tex-204-H
specifications, which proposes preparation of 6 in. diameter and 4.5 in. high specimens using 75
vii
gyrations of the SGC. Furthermore, it is recommended to cure specimens for 24 hrs at room
temperature (77°F) before conducting the indirect tensile strength because the results from this
procedure were more sensitive to asphalt content while reducing the mix design period. The
appropriate asphalt content should satisfy a target indirect tensile strength of at least 85 psi, and a
relative density of 97%.
viii
Table of Contents
Acknowledgements..................................................................................................v
Abstract .................................................................................................................. vi
Table of Contents................................................................................................. viii
List of Tables ...........................................................................................................x
List of Figures ....................................................................................................... xii
List of Figures ....................................................................................................... xii
Chapter One: Introduction .......................................................................................1
1.1 Objectives and Scope.............................................................................2
1.2 Organization...........................................................................................4
Chapter Two: Review of Literature .........................................................................5
2.1 Compaction Methods .............................................................................6
2.2 TxDOT Specifications .........................................................................11
2.3 Other DOTs Current Specifications.....................................................16
2.4 International Review............................................................................19
2.5 Performance of ATB............................................................................20
2.6 Cost-Benefit .........................................................................................26
Chapter Three: District Survey ..............................................................................29
Chapter Four: Comprehensive Evaluation of Alternative Protocols .....................36
4.1 Tex-126-E ............................................................................................36
4.2 Tex-204-F ............................................................................................37
4.3 Proposed Tex-126-H............................................................................37
4.4 Proposed Tex-204-H............................................................................37
4.5 Comparison of Properties from Different Approaches........................38
Chapter Five: Evaluation of Parameters that Impact Performance........................60
5.1 Impact of Number of Gyrations...........................................................60
5.2 Impact of Curing Temperature.............................................................66
5.3 Impact of Binder Grade .......................................................................73
ix
5.4 Impact of AC Content..........................................................................79
5.5 Impact of Gradation .............................................................................81
Chapter Six: Proposed Mix Design Protocol .........................................................89
6.1 Best Mix Design Selection...................................................................90
6.2 Determination of OAC.........................................................................92
Chapter Seven: Field Performance Monitoring .....................................................95
7.1 Laboratory Results ...............................................................................95
7.2 Plant-Mixed Materials .........................................................................99
7.3 Field Cores .........................................................................................102
7.4 Field Results.......................................................................................105
Chapter Eight: Conclusions and Recommendations............................................110
References............................................................................................................112
Curriculum Vita ...................................................................................................114
x
List of Tables
Table 2.1: Mix Requirements for ATB as per Item 292. .............................................................. 12
Table 2.2: Aggregate Quality Requirements. ............................................................................... 13
Table 2.3: Minimum Size of Samples as per Tex 204 Part III. .................................................... 14
Table 2.4: Material, Mixing, and Compacting Temperatures as per Tex 204 Part III.................. 14
Table 2.5: Compaction Parameters as per Tex 204 Part III. ......................................................... 15
Table 2.6: Aggregate Mix Requirements...................................................................................... 16
Table 2.7: Aggregate Base Gradations in Alaska. ........................................................................ 17
Table 2.8: Mix Requirements for ATB Design in Arkansas. ....................................................... 18
Table 2.9: Aggregate Requirements. ............................................................................................ 19
Table 2.10: Precision Estimates.................................................................................................... 27
Table 3.1: Districts that have used ATB in their construction based on Item 292. ...................... 29
Table 3.2: Districts that have used ATB in their construction based on Item 345. ...................... 30
Table 3.3: Candidate Districts Considered for This Study. .......................................................... 35
Table 4.1: Summary of Mix Design Methods. ............................................................................. 36
Table 4.2: Soil Classification and Plasticity Index for Bases under Study................................... 40
Table 4.3: Sand Equivalency and Wet Ball Mill and Hardness for Bases under Study. .............. 40
Table 4.4: Density, Strength and Modulus at Different Asphalt Contents as per Tex-126-E. ..... 43
Table 4.5: Optimum Asphalt Content based on Density, Strength or Modulus as per Tex-126-E.
....................................................................................................................................................... 43
Table 4.6: Density Strength and Modulus at Different Asphalt Contents as per Tex-126-H....... 45
Table 4.7: Optimum Asphalt Contents Based on Density, Strength or Modulus as per Tex-126-H.
....................................................................................................................................................... 45
xi
Table 4.8: Density, Strength and Modulus at Different Asphalt Contents as per Tex-204-H...... 47
Table 4.9: Optimum Asphalt Contents Based on Density, Strength or Modulus as per Tex-204-H.
....................................................................................................................................................... 48
Table 4.10: Maximum Theoretical Specific Gravities for Tex-204-H Specimens. ...................... 49
Table 4.11: Bulk Specific Gravities for Tex-204-H Specimens. .................................................. 49
Table 4.12: Volumetric Properties for Tex-204-H Specimens. .................................................... 51
Table 4.13: Hamburg Wheel Test Results for All Materials. ....................................................... 51
Table 4.14: Impact of Tube Suction Tests on UCS and Seismic Modulus................................... 53
Table 5.1: Sensitivity of Density to Asphalt Content for Tex-126-H Protocol. ........................... 65
Table 5.2: Sensitivity of Density to Asphalt Content for Tex-204-H Protocol. ........................... 69
Table 5.3: Original and Modified Gradation. ............................................................................... 83
Table 6.1: Summary of Mix Design Methods Studied. ................................................................ 89
Table 6.2: MD Curve Sensitivity Results. .................................................................................... 91
Table 7.1: Sand Equivalency, Wet Ball Mill and Hardness for Materials Used. ......................... 96
Table 7.2: Variations in Modulus with Asphalt Content for Alpine, Beaumont and Houston
Materials. ...................................................................................................................................... 99
Table 7.3: Asphalt Contents of Plant-Mixed Materials. ............................................................... 99
Table 7.4: Bulk and Maximum Theoretical Specific Gravities for Plant-Mixed Materials Molded
at 75 Gyrations............................................................................................................................ 101
Table 7.5: Properties of Plant-Mixed Molded using 75 Gyrations of SGC................................ 102
Table 7.6: Delivery and Compaction Temperatures of Asphalt Material at the Sites. ............... 103
Table 7.7: Volumetric Properties of Field Cores. ....................................................................... 103
Table 7.8: Strength and Modulus obtained from Field Cores..................................................... 105
xii
List of Figures
Figure 2.1: Kneading Compactor.....................................................................................................9
Figure 2.2: Hveem Machine used for Hveem Mix Design Method. .............................................10
Figure 2.3: Marshall Hammer........................................................................................................10
Figure 2.4: Marshall Stability. .......................................................................................................11
Figure 3.1: Specifications Used for ATB Design. .........................................................................31
Figure 3.2: Compactors Used for ATB Design. ............................................................................31
Figure 3.3: Main Uses of ATB. .....................................................................................................32
Figure 3.4: Factors for Selection of ATB in Projects. ...................................................................33
Figure 3.5: Binder Used for ATB Projects. ...................................................................................34
Figure 4.1: Gradation Curves from Different Materials Used in This Study. ...............................38
Figure 4.2: Material Constituents for the Sites Being Studied. .....................................................39
Figure 4.3: Density/Modulus /Strength vs. Asphalt Content as per Tex-126-E. ...........................42
Figure 4.4: Density/Modulus /Strength vs. Asphalt Content as per Tex-126-H............................44
Figure 4.5: Density/Modulus /Strength vs. Asphalt Content as per Tex-204-H............................46
Figure 4.6: Gradation Curves from Different Materials Compared to Item 344 Grade SP-A.......48
Figure 4.7: Volumetric Properties for El Paso Material. ...............................................................50
Figure 4.8: Comparison of Optimum Asphalt Contents Based on Density, Strength, and
Modulus. ........................................................................................................................................54
Figure 4.9: Comparison of Tex-126-E OAC with other Methods Based on Density....................55
Figure 4.10: Comparison of Tex-126-E OAC with other Methods Based on Strength.................55
Figure 4.11: Comparison of Tex-126-E OAC with other Methods Based on Modulus. ...............56
Figure 4.12: Densities at OACs for Each Test Method. ................................................................57
xiii
Figure 4.13: UCS at OACs for Each Test Method. .......................................................................57
Figure 4.14: IDT at OACs for Each Test Method. ........................................................................58
Figure 4.15: Seismic Modulus at OACs for Each Test Method. ...................................................59
Figure 5.1: Impact of Number of Gyrations on OAC Based on Density for Tex-126 –H Protocol.61
Figure 5.2: Impact of Number of Gyrations on Density for Tex-126-H Protocol. ........................61
Figure 5.3: Impact of Number of Gyrations on UCS for Tex-126-H Protocol..............................62
Figure 5.4: Impact of Number of Gyrations on Modulus for Tex-126-H Protocol. ......................62
Figure 5.5: Locking Points for Tex-126-H and Tex-126-E Specimens.........................................63
Figure 5.6: Comparison of Density from Tex-126-E and Tex-126-H Protocols. ..........................64
Figure 5.7: Impact of Number of Gyrations on OAC for Tex-204-H Protocol. ............................67
Figure 5.8: Impact of Number of Gyrations on Density for Tex-204-H Protocol. ........................67
Figure 5.9: Impact of Number of Gyrations on IDT Strength for Tex-204-H Protocol. ...............68
Figure 5.10: Impact of Number of Gyrations on Modulus for Tex-204-H Protocol. ....................68
Figure 5.11: Comparison of Density from Tex-126-E and Tex-204-H Protocols. ........................69
Figure 5.12: Impact of Curing Temperature on UCS for Tex-126-H Protocol. ............................71
Figure 5.13: Impact of Curing Temperature on Modulus for Tex-126-H Protocol.......................71
Figure 5.14: Comparison of Strengths from Different Curing Regimes using Tex-126-H
Specimens with Those from Tex-126-E. .......................................................................................72
Figure 5.15: Comparison of Strengths from Different Curing Regimes using Tex-126-E
Specimens. .....................................................................................................................................72
Figure 5.16: Impact of Curing Temperature on IDT Strength for Tex-204-H Protocol................74
Figure 5.17: Impact of Curing Temperature on Modulus for Tex-204-H Protocol.......................74
xiv
Figure 5.18: Comparison of Strengths from Different Curing Regimes using Tex-204-H
Specimens. .....................................................................................................................................75
Figure 5.19: Impact of Binder Grade on OAC based on Density for Tex-126-H Protocol. ..........75
Figure 5.20: Impact of Binder Grade on Density for Tex-126-H Protocol. ..................................76
Figure 5.21: Impact of Binder Grade on UCS for Tex-126-H Protocol. .......................................76
Figure 5.22: Impact of Binder Grade on OAC for Tex-204-H Protocol. ......................................77
Figure 5.23: Impact of Binder Grade on Density for Tex-204-H Protocol. ..................................78
Figure 5.24: Impact of Binder Grade on IDT Strength for Tex-204-H Protocol...........................78
Figure 5.25: Impact of Binder Grade on Modulus for Tex-204-H Protocol..................................79
Figure 5.26: Impact of Asphalt Content Variation on UCS for Tex-126-H Protocol....................80
Figure 5.27: Impact of Asphalt Content Variation on Modulus for Tex-126-H Protocol. ............80
Figure 5.28: Impact of Asphalt Content Variation on IDT for Tex-204-H Protocol.....................81
Figure 5.29: Impact of Asphalt Content Variation on Modulus for Tex-204-H Protocol. ............82
Figure 5.30: Impact of Fines Content Variation on OAC Tex-204-H Protocol. ...........................84
Figure 5.31: Impact of Fines Content Variation on Density Tex-204-H Protocol. .......................84
Figure 5.32: Impact of Fines Content Variation on UCS Tex-126-H Protocol. ............................85
Figure 5.33: Impact of Fines Content Variation on Modulus for Tex-126-H Protocol. ................85
Figure 5.34: Impact of Fines Content Variation on OAC for Tex – 204 –H.................................86
Figure 5.35: Impact of Fines Content Variation on Density for Tex – 204 –H.............................86
Figure 5.36: Impact of Fines Content Variation on IDT for Tex – 204 –H...................................87
Figure 5.37: Impact of Fines Content Variation on Modulus for Tex – 204 –H. ..........................88
Figure 6.1: Density/IDT Asphalt Content Combined Curves........................................................93
Figure 6.2: Relative Density/IDT Asphalt Content Combined Curves. ........................................93
xv
Figure 6.3: Relative Density/IDT Asphalt Content Combined Curves showing Allowable Asphalt
Content Zone based on Density. ....................................................................................................94
Figure 6.4: Relative Density/IDT Asphalt Content Combined Curves showing Allowable Asphalt
Content Zones based on both Density and IDT. ............................................................................94
Figure 7.1: Locations of Sites Visited used in this Study. .............................................................97
Figure 7.2: Pavement Profiles for Sites Visited in this Study........................................................98
Figure 7.3: Laboratory Gradation Curves for Alpine, Beaumont and Houston Materials. ...........98
Figure 7.4: Density/ Strength vs. Asphalt Content for Alpine Material. .....................................100
Figure 7.5: Density/ Strength vs. Asphalt Content for Houston Material. ..................................100
Figure 7.6: Density/ Strength vs. Asphalt Content for Beaumont Material.................................100
Figure 7.7: Gradation Curves from Plant-Mixed Materials as Delivered....................................104
Figure 7.8: PSPA and FWD from ATB Sites. .............................................................................106
Figure 7.9: Comparison between Air Voids of Laboratory and Field Specimens.......................107
Figure 7.10: Comparison between Densities of Laboratory and Field Specimens......................107
Figure 7.11: IDT Comparison between Laboratory and Field Specimens. .................................108
Figure 7.12: Modulus Comparison between Laboratory and Field Specimens...........................109
1
Chapter One: Introduction
One of the alternative stabilized bases that are available to TxDOT districts is the asphalt-
treated bases (ATB) that falls under Item 292 of current specifications. According to the
National Asphalt Pavement Association (NAPA), ATB is a dense-graded HMA with a wide
gradation band and lower asphalt content (2.5% - 4.5%) intended for use as a base course. Also
according to NAPA, ATB should costs less than typical HMA mixes because it can be produced
with less expensive aggregates and lower percentages of asphalt binder. ATB can also provide a
waterproof barrier to prevent fines infiltration into the subgrade and pavement structure and an
alternative to untreated base material.
A review of the specifications of all fifty highway agencies that are incorporated in the
FHWA National Highway Specifications indicates that most agencies either favor designing the
ATB’s using the HMA specifications, or using emulsion instead of asphalt. Moreover, most mix
designs are carried out by using Superpave Gyratory Compactor (SGC), the Hveem method or
the Marshall method. Similarly, newest version of FAA Item P-403 has reverted to Marshall
mix design with the additional requirements of the Tensile Strength Ratio (TSR) from ASTM D
4867.
Current TxDOT construction activities have pointed out that about half a dozen districts
place ATB’s, with the Houston and Beaumont Districts being by far the most frequent users.
However, most districts utilize Tex-204-F to achieve their mix designs primarily because the
compactor specified in Item 292 is not available to all districts. Therefore, an updated mix
design for ATB is in high need. To develop modern test protocols for designing ATB’s, the first
consideration is to determine whether the ATB should be designed and used as a high-quality
base (similar to stabilized or emulsion bases) or as a lower quality hot mix (as compared to
Types A and B asphalt mixes). The mix design requirements for these two alternatives are
different, which in turn will impact the mix design process from compaction of specimens to
their performance testing. No matter which alternative (high-quality base or low-quality hot
2
mix) is pursued, for the ease of operation, the mix design should be compatible with current
practices of TxDOT to minimize the learning curve for the technicians and the acquisition of
new equipment.
One major criterion for the new mix design should be the compatibility between the
parameters measured in the lab and the similar parameters obtained from the as-constructed
layers. Currently, quality management of ATB’s during construction is for the most part is based
on the measured density under Tex-207-F using a nuclear density gauge on the field, or in the lab
on cores or prepared samples retrieved from the site. There is some concern that the densities
measured in those fashions may not be accurate or repeatable because of the nature of the base
material (limited control on the gradation) used in ATB’s. Since the accurate measurement of
the maximum theoretical density of the mix measured as per Tex-227-F directly impacts the
relative density of the ATB, the accuracy and precision of this test for use with the ATB needs
also to be evaluated.
Based on this background, our goals in this project are to achieve the following items:
• Define criteria and procedures that are compatible with production and placement
requirements of ATB,
• Develop a new simple, efficient, and consistent mix design method that uses
commercially available equipment.
• Develop draft laboratory specification requirements for better quality control of
asphalt treated base.
1.1 OBJECTIVES AND SCOPE
The main goal of this project was to develop a laboratory test protocol to help in selecting
the optimum asphalt content and a guideline or draft specification for the construction of asphalt-
treated bases. To achieve this goal, the following objectives were addressed:
1. Document different uses and establish the most appropriate uses of the ATB in
Texas considering both the engineering and economical consideration.
3
2. Evaluate the reasonableness, strengths and weaknesses of current practices (e.g.
Item 292, Tex-126-E, and Tex-204-F) in terms of mix design and construction
practices.
3. Evaluate and establish the most appropriate performance indicators and ways of
designing ATB’s in the laboratory based on the intended use and identified
performance indicators.
4. Evaluate the best method of compacting the specimens given the existing
available equipment in TxDOT such as the Superpave Gyratory Compactor
(SGC) instead of the Texas gyratory compactor (TGC) specified in Item 292.
5. Develop the compaction criteria in terms of energy and/or number of gyrations
that is more representative of this type of mixes in the field.
6. Evaluate, and if necessary modify, the process of determining the optimum
asphalt content for these mixes from the process enumerated in Tex-126-E or
Tex-204-F.
7. Establish the minimum strength characteristics of ATB and the best way to
measure them so that the value-added benefits of the binder added to the mix is
best represented.
8. Incorporate tests (e.g. moisture susceptibility) to ensure long-term performance of
the mix.
9. Evaluate and recommend the best construction practices with especial attention to
the reasonableness and accuracy of the current quality management process.
10. Verify the mix design and the results from laboratory testing by evaluating the
field performance.
11. Monitor and record the initial performance of pavement sections with ATB.
4
1.2 ORGANIZATION
Chapter Two contains a thorough literature review of studies addressing current
compaction methods; specifications used in Texas and many other states and countries;
performance of Asphalt Treated Base (ATB), and cost benefit of the use of the ATBs.
Chapter Three presents the result of district survey conducted at the beginning of this
research. The districts that previously have used ATB were identified. Construction and
laboratory specifications and compactors used by those districts were as well identified. The
survey also collected the main uses of ATB, factor to motivate its use, type of binder grade and
aggregate, and criteria and problems found during design or construction of ATB.
Chapter Four reflects the work done at the early middle of the project in order to compare
current (Tex-126-E and Tex-204-F) and proposed (Tex-126-H, Tex-204-H) protocols. Materials
were retrieved from different districts to prepare specimens using both: Texas and Superpave
Gyratory compactors; the results were analyzed and compared. The properties of each material
obtained were also addressed.
Chapter Five contains a detailed evaluation of the parameters that impact the performance
of a mix for the proposed mix design protocols. The measured impact of the number of
gyrations, curing temperature, binder grade, change in gradation, and asphalt content variation to
the property of the mixes for the proposed procedures were addressed.
Chapter Six describes the mix design protocol selected. The compactor, specimen size,
density calculation, number of gyrations determined to meet the new protocol requirements are
presented. Strength parameter, curing and testing temperature, density and strength requirements
and the new method for determining the optimum asphalt content are also showed.
Chapter Seven contains the information obtained from field investigation conducted at
several sites in order to evaluate the results of the proposed protocol. The laboratory results and
properties of the materials acquired from every field sites are also discussed.
5
Chapter Two: Review of Literature
The performance of a pavement depends on many factors such as the properties of the
materials used, structural capacity of the pavement, construction method, traffic loading, and
climatic conditions. For flexible pavements, the quality of the base layer is one of the most
important factors. Previous research has found that much of the distress that flexible pavements
experience can be traced to problems encountered in the base (Saeed et al., 2001). The use of a
cost efficient base layer, that would extend the pavement life, that would require less thickness,
or that would use local materials, is highly desirable. Asphalt-treated bases (ATB) fit this
category.
According to the National Asphalt Pavement Association (NAPA),
(http://training.ce.washington.edu/wsdot/Modules/02_pavement_types/02-3_body.htm), ATB is
a dense-graded hot mix asphalt (HMA) with a wide gradation band with a lower asphalt content
that can be used as a base course. Among the features that make it different from HMA are
(Wong et al., 2004):
1. HMA layer may receive direct impact from traffic and experience more serious
weather conditions than those experienced by the ATB
2. Thickness of ATB is greater than the one of the HMA
3. Asphalt layer can be superimposed to improve surface performance, or removed
to construct a new surface layer when the damage has attained certain level
These differences make the ATB’s design vital, since it must preserve necessary
performance in the entire pavement life.
One of the main problems with asphaltic pavements that include a stabilized base course
is the incidence of transverse cracking in the stabilized base and the related propagation of cracks
to the surface, which diminishes the life of pavements. In this case, ATB may be more flexible
and resistant to fatigue cracking as compared to cement stabilized bases (Dykman et al., 2003).
6
Besides new construction, ATB can be beneficial in rehabilitation projects. According to
Dykman et al. (2003), the crucial factors in choosing ATB as a way of rehabilitation are:
• Quick construction time, therefore decreased delays and diminished traffic
disturbance
• Low permeability to moderate pore water pressure effect (Cedergren, 1977)
• Minimal moisture sensitivity as a consequence of moisture ingress
• Quite flexible, consequently decrease the possibility of reflective cracking (Marks
and Heisman, 1985.)
Research in ATBs has been limited even though ATBs have been used as structural
pavement layers for more than 40 years (Dykman et al., 2003). McDowell and Smith (1969)
performed the first comprehensive study in design and construction of ATB (also known as black
base). An important objective of their study was to observe the effects of rate of loading on the
unconfined compressive strength (UCS) of ATB. McDowell and Smith used loading rates of 6,
8, 10 and 15 in./min, noticing that when testing at fast rates of loading a definite improvement in
compressive strength of asphalt treated materials over untreated materials was obtained. They
stated that a fast rate of loading test should become part of the analysis of asphalt mixtures.
McDowell and Smith (1969) also studied the effects of moisture absorption on strength
and its relation to total percent voids. They used pressure pycnometer to obtain saturation in the
least amount of time. They concluded that mixtures having less than 5.5% total voids will
possibly not lose strength due to absorption of moisture. While McDowell and Smith
investigation was taking place, the Texas Gyratory Compactor (TGC) was revised so that 6 in. in
diameter by 8 in. in height specimens could be prepared.
2.1 COMPACTION METHODS
An analysis of the specifications of all fifty highway agencies that are incorporated in the
Federal Highway Administration (FHWA) National Highway Specifications indicates that most
highway agencies either support designing the ATBs using the HMA specification (e.g., Illinois,
7
Indiana, Washington), or utilizing emulsion rather than asphalt (e.g., Delaware, Maine,
Maryland). However, several transportation agencies are upgrading their conventional
compaction methods such as TGC or Marshall with SGC for routine mix design. According to
Button et al. (2006) the advantages of using the SGC include the following items:
• Its ability to estimate the compatibility of mixes from density during the
compaction process
• Its ability to identify weak aggregate structures that collapse very quickly to lower
air voids
• Improved reproducibility of samples due to mostly mechanical control of
compaction process
• Ability to simulate field compacted mixes relatively better than other compaction
methods
Gyratory compaction was originally created in 1939 by the Texas Highway Department
to help in the molding and design of asphalt mixtures (Harman et al., 2002). According to
Sebesta et al. (2004), a more advanced Texas Gyratory Compactor has also been used to compact
soil samples which were subsequently used in laboratory swell tests.
Gyratory compactors were designed to simulate the orientation of aggregate, degradation
of aggregate, field compaction, and traffic degradation that occurs in HMA during production,
compaction and traffic loading (Collins et al., 1997). Dykman et al. (2003) state that gyratory
compaction can simulate the action of a roller in the field due to its capability to rotate the
principal stresses. On the other hand, gyratory compaction can create totally different
compaction characteristics. These characteristics depend on the adjustment and calibration of
several parameters that affect the degree of compaction of laboratory HMA specimens (Mokwa
et al., 2008).
According to Button et al. (2006), the lower angle of gyration of the SGC (1.25°) imparts
significantly less mechanical energy into the specimen per gyration as compared to the TGC
(5.8° gyration angle). Different angles of gyration have different influence on the orientation of
8
the aggregates, particularly the larger aggregates. The differences between specimens prepared
using the TGC and SGC such as: air void structure, aggregate orientation, voids in mineral
aggregate (VMA) and density gradient, will not likely be consistent because these differences
will depend on the shear resistance of the mixture.
Mokwa et al. (2008) used confining pressures ranging from 30 psi to 90 psi in preparing
specimens with a SGC. They found that mixes with smaller particles sizes exhibited higher rates
of densification as a result of higher confining pressures. The confining pressure applied to the
specimen according to Tex-126-E varies from 20 psi to 60 psi for the TGC.
Aguiar-Moya et al. (2007) indicate that the basis for the number of gyrations is that the
compactive effort obtained in the laboratory should produce the same outcome on the asphalt
mixtures (to increase density) as traffic loads for in-place asphalt mixes. As more weight is
assigned to fatigue resistance, the optimum number of gyrations decreases, therefore producing
mixes with higher binder contents. Similarly, as more significance is given to rutting, the
optimal number of gyrations increases. Button et al. (2006) indicate that the design gyrations in
SGC can be reduced below the initial recommendations without compromising rutting resistance
of HMA mixtures.
Although the SGC can produce the same volume of air voids as the TGC in a given
mixture type, the resulting optimum asphalt contents and engineering properties of the
compacted mixtures may be measurably different because of different aggregate orientations and
different density gradients within the specimens (Button et al., 1994; Von Quintus et al., 1991).
According to NAPA,
http://training.ce.washington.edu/wsdot/Modules/05_mix_design/05-3_ body.htm) the Hveem
method has been proven to produce quality HMA from which long-lasting pavements can be
constructed. Hveem method has the following six main steps:
1. Aggregate selection
2. Asphalt binder selection
3. Sample preparation using the California Kneading Compactor (Figure 2.1)
9
4. Stability and cohesion determination using a Stabilometer and Cohesiometer
(Figure 2.2)
5. Density and voids calculations, and
6. Optimum asphalt binder content selection.
The basic concept of the Marshall mix design method was originally developed by Bruce
Marshall of the Mississippi Highway Department around 1939 and then refined by the U.S.
Army. The Marshall method is very popular because of its relatively simplicity, economical
equipment and proven record. Similar to Hveem method, the Marshall method consists of the
following six basic steps:
1. Aggregate selection
2. Asphalt binder selection
3. Sample preparation using the Marshall Hammer (Figure 2.3)
4. Stability and Flow Test using the Marshall Stability testing apparatus (Figure 2.4)
5. Density and voids calculations, and
6. Optimum asphalt binder content selection
Figure 2.1: Kneading Compactor.
10
a) Stabilometer b) Cohesiometer
Figure 2.2: Hveem Machine used for Hveem Mix Design Method.
According to Wong et al. (2004) conventional Marshall method is not appropriate for
ATB mixture design for the reason that the maximum size of aggregate and the thickness of ATB
may not be comparable to those of asphalt layer.
Figure 2.3: Marshall Hammer.
11
Figure 2.4: Marshall Stability.
2.2 TXDOT SPECIFICATIONS
An evaluation of TxDOT construction activities pointed out that about six out of twenty-
five districts place ATBs, with Houston and Beaumont Districts being the most recurrent users.
In Texas, ATBs are traditionally designed and constructed as per Item 292 “Asphalt Treatment
(Plant Mixed),” of the 2004 Standard Specification book. Since the compactor under Item 292 is
not available to all districts, some districts have started using Tex-204-F, Part III ‘Mix Design for
Large Stone Mixtures Using the Superpave Gyratory Compactor.”
2.2.1 Item 292: Asphalt Treatment (Plant-Mixed)
Table 2.1 shows the mix requirements for Item 292. This specification is a hybrid of base
and hot mix asphalt concrete procedures and requirements. Under Item 292, the aggregates
basically have the same gradation and quality as Item 247 for untreated base, which is less
rigorous than those utilized in Type A/B mixes under Item 341 or 344. Aggregate quality
requirements for Item 292 are reflected in Table 2.2. Item 292 permits the use of crushed
concrete in the mix, which is usually unacceptable in the Type A/B mixes. It seems that the
main incentive for the utilization of Item 292, as stated by NAPA, may be to incorporate local
materials (raw or recycled) in the local construction.
12
Table 2.1: Mix Requirements for ATB as per Item 292.
Master Gradation Bands Tex-200-F, Part I, % Passing by Weight
Sieve Size Grade 1 Grade 2 Grade 3 Grade 4
1-3/4" 100 100
1-1/2" 100 90-100
1" 90-100
3/8" 45-70
#4 30-55 25-55
#40 15-30 15-40 15-40
As shown on the
plans
Strength Requirements
Slow strength, psi,
min.1 50 40 30 30 2
1. At optimum asphalt content 2. Unless a higher minimum strength is shown on the plans
Under Item 292, 3% to 9% binder is suggested for the ATB; however, based on our
review of several current mix designs from several districts, the optimum binder content is about
4% to 5%. For Type A/B mixes the optimum binder content is typically 5% to 6%. The
unconfined compressive strength of the mix (as per Tex-126-E) is used to assess the quality of
the mix. Since the placement of the mixes under Item 292 and 341/344 is similar (with less strict
field quality management for Item 292), it seems that ATB cost should be slightly less than Type
A/B mixes.
2.2.2 Tex-126-E: Molding, Testing, and Evaluating Bituminous Black Base Materials
This method is used to mold an asphalt stabilized (black base) material, and to determine
the relationship between asphalt content and density (this relationship is called an asphalt-density
curve) and similarly an unconfined compressive strength-density curve. The compacted black
base specimens are made in duplicates and are tested for their unconfined compressive strengths
13
Table 2.2: Aggregate Quality Requirements.
Property Test Method Specification Requirement
Wet ball mill, % max 50
Max increase, % passing #40 Tex-116-E
20
Liquid Limit, max Tex-104-E 40
Plasticity Index, max Tex-106-E 10
Sand Equivalent, % min Tex-203-F 40
at 140°F. The specimens are subjected to two types of deformation rates: a slow (0.15 in./min)
and a fast deformation rate (10 in./min). Specimens tested at a slow deformation rate yield
relatively lower strengths as compared to the fast deformation rate. From the strength-density
relationship, the minimum density that would satisfy the unconfined strength requirements as per
Item 292 is taken as the minimum allowable density.
2.2.3 Tex 204 Part III: Mix Design for Large Stone Mixtures Using Superpave Gyratory Compactor (SGC)
This procedure was originally developed to design Type A and Type B hot mixes at 96%
density from a 6 in. by 4.5 in. specimens as per Table 2.3. Depending on the PG grade of the
binder, the material, mixing, and compaction temperatures are according to Table 2.4. The
specimens are molded to either 100 gyrations, or as shown on the plans, or as per Table 2.5.
The specimens prepared under Tex-204-F are not the appropriate sizes for unconfined
compressive strength (as per Tex-126-E). As such, the quality of the mix assessed with the
indirect tensile strength at a deformation rate of 2 in./min. at a temperature of 77 ± 2°F according
to Tex-226-F.
14
Table 2.3: Minimum Size of Samples as per Tex 204 Part III.
Nominal Maximum Size of Particles,
Passing Sieve Minimum Weight of Sample for Test, lb
Coarse Aggregate
2" 8
1-1/2" 8
1" 6
3/4" 4
1/2" 3
3/8" 2
Fine Aggregate
#4 1.1
#8 1.1
Table 2.4: Material, Mixing, and Compacting Temperatures as per Tex 204 Part III.
PG Grade Asphalt Material
Temperature, °F
Mixing
Temperature, °F
Compaction
Temperature, °F
64-22 290 290 250
64-28 300 300 275
70-22 300 300 275
70-28 325 325 300
76-16 325 325 300
76-22 325 325 300
15
Table 2.5: Compaction Parameters as per Tex 204 Part III.
Compaction Parameters Design
ESALs1
(million) N initial N des N maximum
Typical Roadway Application 2
<0.3 6 50 75
Applications include roadways with very light
traffic volumes such as local roads, county roads,
and city streets where truck traffic is prohibited or
at a very minimal level. Traffic on these roadways
is local in nature, not regional, intrastate, or
interstate. Special purpose roadways serving
recreational sites or areas may also be applicable to
this level.
0.3 to <3 7 75 115
Applications include many collector roads or
access streets. Medium-trafficked city streets and
the majority of country roadways may be
applicable to this level.
3 to <30 8 100 160
Applications may include many 2-lane, multilane,
divided, and partially or completely controlled
access roadways. Among these are medium to
highly trafficked city streets, many state routes, US
highways, and some rural interstates.
≥30 9 125 205
Applications include the vast majority of the US
Interstate System, both rural and urban in nature.
Special applications such as truck-weighing
stations or truck-climbing lanes on 2-lane
roadways may also be applicable to this level. 1 Design ESALs are the anticipated project traffic level expected on the design lane over a 20-yr. period. Regardless of the actual design life of the roadway, determine the design ESALs for 20 yr., and choose the appropriate Ndes level 2 Typical Roadway Applications as defined by A Policy on Geometric Design of Highway and Streets, AASHTO
16
2.3 OTHER DOTS CURRENT SPECIFICATIONS
As mentioned before some agencies either support designing the ATBs using the HMA
specifications, or utilizing emulsion rather than asphalt. Other agencies specify design processes
for asphalt treated permeable bases, which are out of the scope of this project. However, few
states have specifications for the ATB design.
2.3.1 Alaska
According to Alaska Department of Transportation (AKDOT) the ATBs are designed
under Section 306 “Asphalt Treated Base Course.” The selection of aggregate is mainly
determined by AASHTO requirements presented on Table 2.6. In Alaska, the aggregate
requirements shown on Table 2.7 are dictated by AASHTO T 27/T 11.
Table 2.6: Aggregate Mix Requirements.
Property Base Course Test Method
L.A. Wear, % 50, max. AASHTO T 96
Degradation Value 45, min. ATM 313
Fracture, % 70, min. WAQTC FOP for AASHTO TP 61
Liquid Limit --- WAQTC FOP for AASHTO T 89
Plastic Index 6, max. WAQTC FOP for AASHTO T 90
Sodium Sulfate Loss, % 9, max. (5 cycles) AASHTO T 104
An important point mentioned in the Alaskan specification is the weather limitations,
which states not to place the asphalt mixture on a wet or frozen surface, or when weather
conditions will prevent proper handling, compacting or finishing of the mixture. It also states
not to place the asphalt mixture unless the air temperature is above 40 °F, as measured in the
shade and away from any heat sources.
2.3.2 Arkansas
Arkansas Department of Transportation’s (ARDOT) design of ATB falls under Section
417 “Open Graded Asphalt Base Course.” Besides the size of the aggregates requirements, there
is a criterion for asphalt content for each grade type as shown on Table 2.8.
17
Table 2.7: Aggregate Base Gradations in Alaska.
Gradation Sieve
C-1 D-1
1-1/2" 100 --
1" 70-100 100
3/4" 60-90 70-100
3/8" 45-75 50-80
#4 30-60 35-65
#8 22-52 20-50
#50 8-30 8-30
#200 0-6 0-6
2.3.3 Washington
Washington Department of Transportation (WSDOT) has its own specification for ATB
design. This design has two general requirements:
1. Los Angeles Wear, 500 Rev. → 30% max.
2. Degradation Factor → 15 min.
When the aggregates are mixed within the limits of Table 2.9 and mixed in the laboratory
with the designated grade of asphalt, the mixture shall be capable of meeting the following test
values:
• Stabilometer Value → 30 min.
• Cohesiometer Value → 50 min.
• Modified Lottman Stripping Test → 80% min.
• Sand Equivalent Value → 30 min.
18
Table 2.8: Mix Requirements for ATB Design in Arkansas.
Sieve Size Type 1 Type 2 Type 3 Type 4
3" 100
2 1/2" 95-100
2" 100
1-1/2" 30-70 [±7] 75-90 [±7]
1" 100
3/4" 0-15 [±7] 50-70 [±7] 100 90-100
1/2" 90-100
3/8" 0-2 20-55 [±5]
#4 8-20 [±5] 0-15 [±5] 0-10
#8 0-3 0-5
#100 0-5
Asphalt Content 1.5 - 4.0 1.5 - 4.0 1.5 - 4.0 2.5 - 3.0
Note: The number in brackets is the allowable tolerance from the mix design value
19
Table 2.9: Aggregate Requirements.
Sieve Size Percent Passing
2" 100
1/2" 56-100
1/4" 40-78
#10 22-57
#40 8-32
#200 2.0-9.0
Asphalt Content 2.5 - 4.5
2.4 INTERNATIONAL REVIEW
2.4.1 Australia
A study in Queensland, Australia by Dykman et al. (2003) mentioned that due to the
population growth, more emphasis is being placed on maintenance and rehabilitation processes.
Due to the relatively thin granular base courses used, pavement rehabilitation techniques now
prefer modification than stabilization. This involves using treatments that balance the materials’
strength and ductility in order to produce strong but flexible granular layer so that fatigue
cracking can be minimized. ATB is considered as an alternative for this purpose. Dykman et al.
(2003) used Marshall and gyratory-compacted specimens. ATB mixes appeared to be fairly rut
resistant and retained a high stiffness. This characteristic of ATB may improve the load
distribution capacity of the pavement, leading to more effective protection of the underlying
layers. Dykman et al. (2003) and Ullman and Nolan (1991) conclude that the following benefits
can be obtained from ATB, if used properly:
• ATB can be placed with conventional equipment
• Fast construction
• Less cost than conventional hot-mix asphalt
20
• High stability
• Possibility of using marginal aggregates
• Potential for recycling
2.4.2 Singapore
Wong et al. (2004) stated that ATB has better strength and stability compared to common
macadam base, and has good flexibility and durability. In the cases of high traffic volume and
increasing axle loading on certain roadways, using ATB can be useful in respect to pavement
performance. According to Wong et al., the determination of asphalt content to get the best
performance is through establishing a balance between friction and cohesion.
Wong et al. (2004) also recommended a procedure that considered the variations of the
maximum density, indirect tensile strength and unconfined compressive strength of a blend with
asphalt content to obtain the optimum mix design. Through static creep and fatigue tests, they
demonstrated that mixture using their design method had good resistance to permanent
deformation and fatigue at the bottom of the base. They prepared their specimens using an SGC,
with the number of gyrations being based on the locking point of the mix. As a conclusion from
that study, Wong et al. (2004) recommended the use of compression test and indirect tensile test
for ATB design since these tests are simple and convenient.
2.5 PERFORMANCE OF ATB
Some of the parameters that play an important role in the performance of the ATB as a
system are the structural integrity of the section, the internal stability of the layer, the
environmental conditions, and most importantly the quality of construction.
21
2.5.1 Structural Integrity
The structural integrity of a flexible pavement section is controlled by several parameters.
In most classical structural design programs (such as FPS19), the design thickness of the layers is
(directly or indirectly) estimated based on the criteria that the stresses at the interfaces of the hot
mix and base, and the base and subgrade, are low enough so that the cracking and rutting will not
be an issue. The traffic volume is also a major consideration. For a given traffic condition, the
thicker the layers overlying the base, the thicker the base layer and the stiffer the subgrade are,
the lower the base layer stresses will be. This indicates that not only the quality of base should
be considered, the stiffness of the subgrade, and the thickness of the hot mix should also be
considered. The complex modulus or diameteral resilient modulus tests can be performed for
this purpose.
2.5.2 Internal Stability
The internal stability is defined as the excessive deformation of the base under the load.
This manifests as rutting primarily associated with the base layer. To address this issue, repeated
load permanent deformation lab tests as advocated by the FHWA should be used in conjunction
with the appropriate models that predicts the rutting of the hot-mix, base, and subgrade layers
individually.
2.5.3 Environmental Conditions
The main environmental parameter of interest is the adverse effects of moisture and
temperature on the strength and modulus of the base. Therefore, the importance of considering
the impact of moisture on the performance of the material should not be neglected. Hamburg
wheel tracking device, perhaps with some relaxed requirements, can be used for this purpose.
Alternatively, two inter-related methods can be used to assess the impact of moisture on the
performance of the base: Tube Suction Test (Tex-145) and the Free-Free Resonant Column
(proposed Tex-147) or V-meter (proposed Tex-259). The Tube Suction Test (TST) qualitatively
provides an estimate of the water-retention of the base material that can be correlated to the
22
potential of damage to the base due to softening. The Free-Free Resonant Column (FFRC) or V-
meter test is a quantitative nondestructive lab method that can be performed on a specimen for its
modulus. In both methods, each specimen is oven-dried for two days and then allowed to soak
moisture through capillary saturation. The modulus of the specimen is measured every day in
conjunction with the tube suction test. The residual modulus corresponds to the modulus
measured after the specimen soaked moisture for several days. Since the same specimen is used
throughout for both TST and FFRC tests, the variation in moisture content with time can also be
obtained by weighing the specimens daily. In that manner, the moisture retention properties of
the material and its impact on modulus can be measured. Of course separate specimens should
be prepared and tested for strength and modulus at optimum and after moisture conditioning to
determine the retained strength and retained modulus for conventional design.
2.5.4 Quality of Construction
No matter how much attention is focused on the mix design, the performance of the mix
is directly related to the quality of construction. The necessity to perform a thorough evaluation
of the component materials, and a thorough testing regimen and an aggressive quality control/
quality assurance program is well understood by TxDOT and is incorporated in the appropriate
specifications. One of the major quality management tool used is the density of the in-place mat.
To successfully assess the density of the mat, two parameters are necessary: The theoretical
maximum specific gravity of the mixture (Tex-227-F) and the bulk specific gravity of the cores
obtained from the finished mat (or density with the NDG) as per Tex-207-F. Both of these
methods have been the subject of numerous studies by the federal and state highway agencies.
There is some concern that the densities measured in those fashions may not be accurate
or repeatable because of the nature of the base material (limited control on the gradation) used in
ATBs. The most recent and comprehensive study of this matter has been carried out under a
multi-year, multi-phase project (NCHRP 9-26) that was completed in 2007 by the AASHTO
Materials Reference Laboratory (AMRL). The report from Phase 1 of Project included precision
23
estimates of selected volumetric properties of HMA using non-absorptive aggregates by
(Spellberg and Savage, 2003). The report from Phase 2 discusses the results of an investigation
into the cause of variations in HMA bulk specific gravity test results using non-absorptive
aggregates by Spellberg and Savage (2004). The report from Phase 3 includes a robust technique
developed by AMRL for analyzing proficiency sample data for the purpose of obtaining reliable
single-operator and multi-laboratory estimates of precision (Holsinger et al., 2005). According
to Azari et al. (2006) the report from Phase 4 includes the precision estimates of selected
volumetric properties of HMA using absorptive aggregates, and the effect of aging period on the
volumetric properties of the absorptive aggregates. The Phase 5 report includes the update
precision estimates for AASHTO Standard Test Method T269, “Percent Air Voids in Compacted
Dense and Open Asphalt Mixtures.” These reports are summarized below.
The inter-laboratory repeatability and reproducibility estimates of several common
volumetric properties were established. A coarse (maximum aggregate size of 19.0 mm) mix
and a fine mix (maximum aggregate size of 12.5 mm) were considered. The following two
different methods were considered for the theoretical maximum specific gravity:
• ASTM D6857 “Specific Gravity and Density of Bituminous Paving Mixtures
Using Automatic Vacuum Sealing Method” and
• ASTM D2041 (equivalent to Tex-227-F Part II) “Standard Test Method for
Theoretical Maximum Specific Gravity and Density of Bituminous Paving
Mixtures.”
Based on the results from 26 laboratories, the values obtained from ASTM D6857 were
found to be lower than those obtained following ASTM D2041 (equivalent to Tex-227-F Part II).
The author mentioned that the variability in ASTM D6857 can be attributed to some problems
while performing the test like: the bags touching the sides of the bath during the weighing, or an
incomplete evacuation of air during sealing process. This variability consequently produced a
statistically significant difference in the repeatability and reproducibility of both tests.
The bulk specific gravity was also measured using the following two other methods:
24
• ASTM D6752 (equivalent to Tex-207-F Part VI) “Bulk Specific Gravity and
Density of Compacted Bituminous Mixtures Using Automatic Vacuum Sealing
Method” and
• AASHTO T166 (equivalent to Tex-207-F Part I) “Bulk Specific Gravity of
Compacted Asphalt Mixtures Using Saturated Surface-Dry Specimens.”
The specimens tested under ASTM D6752 (equivalent to Tex-207-F Part VI) had a lower
average specific gravity as determined from tests performed using AASHTO T166 (equivalent to
Tex-207-F Part I). Using student T-test statistical analyses, there was a good reproducibility for
both coarse and fine specimens tested under AASHTO T166 (equivalent to Tex-207-F Part I);
however the F-test statistical analyses showed a statistically significant difference between the
repeatability of both mixes. Variability was attributed to incomplete evacuation of air during
sealing process and pinholes developing in the bags after the vacuum sealing process. This study
also stated that there was a significant difference between the densities of different compactors.
For example, the specimens compacted with a Troxler 4140 compactor had a lower density than
the specimens compacted using a Pine AFGC125X. The results of this inter-laboratory study
also indicated that the within laboratory variation in the bulk specific gravity test results was
much greater than the variation in the maximum theoretical specific gravity test results.
The second phase of this study focused on the variability of the bulk specific gravity.
Instead of each laboratory compacting its own specimens as done in Phase I, the necessary
specimens were produced at the same facility to minimize the compaction errors. Data from 21
laboratories, each one testing specimens with maximum size of 9.5-mm, 12.5-mm, and 19.0-mm
showed that overall the AASHTO T166 (equivalent to Tex-207-F Part I) measurements were
precise. Comparing data from Phases I and II, the authors conclude that 90% of the variation in
AASHTO T166 (equivalent to Tex-207-F Part I) bulk density test results can be attributed to the
mixing and compaction process.
The next phase of this inter-laboratory study focused on the development of precision
statements of absorptive aggregates applicable to the following tests:
25
• AASHTO T312 (equivalent to Tex-241-F Part I) “Preparing and Determining
Density of HMA Specimens by Means of the SGC”,
• AASHTO T166 (equivalent to Tex-207-F Part I) “Bulk Specific Gravity of
Compacted Asphalt Mixtures Using Saturated Surface-Dry Specimens”,
• ASTM D2041 (equivalent to Tex-227-F Part II) “Theoretical Maximum Specific
Gravity and Density of Bituminous Paving Mixtures”, and
• ASTM D6752 (equivalent to Tex-207-F Part VI) “Bulk Specific Gravity and
Density of Compacted Bituminous Mixtures Using Automatic Vacuum Sealing
Method”
According to the authors, the precision statement for AASHTO T312 (equivalent to Tex-
241-F Part I) should differ using absorptive aggregates and non-absorptive aggregates based on
the significant difference between repeatability and reproducibility estimates for relative density
measurements obtained in this study. The conclusion of the authors was that the repeatability
and reproducibility estimates of the theoretical maximum specific gravity are significantly
different using absorptive and non-absorptive aggregates. They suggested following ASTM
D2041 Section 11 “Supplemental Procedure for Mixtures Containing Porous Aggregate”
(equivalent to Tex-227-F Part II) for absorptive aggregates mixtures.
For Bulk Specific Gravity AASHTO T166 (equivalent to Tex-207-F Part I) and ASTM
D6752 (equivalent to Tex-207-F Part VI) were compared. The average specific gravity obtained
using ASTM D6752 (equivalent to Tex-207-F Part VI) was significantly lower than the average
specific gravity determined by AASHTO T166. However, the repeatability and reproducibility
estimates of both tests were comparable for absorptive and non-absorptive mixtures. The authors
also stated that it was appropriate to combine repeatability and reproducibility estimates of
ASTM D6752 (equivalent to Tex-207-F Part VI) and AASHTO T166 (equivalent to Tex-207-F
Part I) for coarse and fine mixtures.
Another objective of this phase was to examine the effect of aging time on bulk specific
gravity, maximum specific gravity, percent air voids, and percent absorbed asphalt of mixtures
26
with absorptive aggregates. Mixtures with more absorptive aggregates were found to be more
susceptible to aging. Since the increase in aging time would allow more absorption and
additional hardening of asphalt binder, this would reduce the compactability and consequently
increase of percent air voids in the compacted mixtures. They recommended that the laboratory
aging time should not exceed 4 hrs to prevent a significant change in the original mixture design.
The final phase of the study consisted of updating the estimates for AASHTO Standard
Test Method T269 “Percent Air Voids in Compacted Dense and Open Asphalt Mixtures.” The
specimens used in that study were compacted by means of a Marshall apparatus, a California
Kneading Compactor, a Gyratory Shear Compactor, and ta Superpave Gyratory Compactor. The
authors recommended the precision limits for AASHTO T269 to be determined based on
precision of percent air voids since this parameter is considered the controlling property for
design and quality control of asphalt mixtures. Based on this study the Marshall apparatus and
Superpave Gyratory Compactor precision estimates were in good agreement with the previous
estimates presented in AASHTO T269 Section 8.3 “Criteria for Judging the Acceptability of
Percent Air Voids Test Results that are obtained by Using T166 and T209 for Non Porous
Aggregates”. On the other hand, precision estimates computed using the California Kneading
and Shear Gyratory Compactors were significantly larger than the ones previously stated on
Section 8.3. Table 2.10 shows the precision estimates of this study based on the repeatability
and reproducibility of the data.
2.6 COST-BENEFIT
According to NAPA, ATB should cost less than typical HMA mixes since it can be
produced with less expensive aggregates and lower percentages of asphalt binder. Among the
advantages of ATB, as stated by NAPA, on a site that must export material (excess cut), an ATB
pavement can save a considerable amount of excavation, hauling and disposal costs.
Furthermore, on a site that must import material (excess fill), ATB can be used to build the
pavement over more marginal subgrades. Another benefit from ATB is that it provides a water
27
proof barrier to prevent fines infiltration into the subgrade and pavement structure. According to
NAPA, if water accumulates in the subgrade, the repetition of pavement loading may cause
subgrade fines to migrate into the base and pavement structure. This can clog the base layer,
which blocks drainage and create voids in the subgrade into which the pavement may settle. One
additional valuable characteristic of ATB is the alternative to untreated base material. According
to NAPA, ATB is structurally up to three times as strong as an untreated granular base. NAPA
states that it may be feasible to use thinner layers for the same structural support.
Table 2.10: Precision Estimates.
Compaction Method Specimen
Diameter, in
Standard
Deviation (1s)a
Acceptable
Ranged of Two
Test Results (d2s)a
Single Operator Precision:
Marshall Apparatusb 4 0.48 1.36
California Kneading Compactor 4 0.52 1.47
Gyratory Shear Compactor 4 0.50 1.42
Superpave Gyratory Compactor 6 0.47 1.33
Multilaboratory Precision:
Marshall Apparatusb 4 1.08 3.06
California Kneading Compactor 4 1.39 3.94
Gyratory Shear Compactor 4 1.49 4.22
Superpave Gyratory Compactor 6 1.01 2.86 a These values represent the 1s and d2s limits described in ASTM Practice C670. b The results reported for specimens compacted using T245 were determined as the average of three specimens. Note: The precision estimates given in the table are based on he analysis of test results from three pairs of AMRL proficiency samples. The data analyzed consisted of results from 20 to 578 laboratories for each of the three pairs of samples. The analysis included three binder grades: PG 70-22, PG 64-10, PG 64-22. Average results for air voids ranged from 2.37% to 7.95%. The details of this analysis are in NCHRP Final Report, NCHRP Project No. 9-26, Phase V.
28
In accordance with feedback from local authorities in Queensland, Australia, ATBs have
performed very well over a period of ten years and in particular, its value for money when
compared to other base stabilization treatments (Dykman et al.,2003). According to McDowell
and Smith (1969) the economic benefit of ATB over other hot mixes is generally dependent on
the use of local base materials, which do not have to be washed and sieved before batching.
29
Chapter Three: District Survey
A survey was conducted to identify the activities related to the use of Asphalt Treated
Base (ATB) throughout Texas, and to identify possible sites to be incorporated in this study. The
questionnaire is included in the appendix at the end of this report.
Survey responses were received from the following 18 districts: Abilene, Amarillo,
Atlanta, Austin, Brownwood, Bryan, Childress, Dallas, Fort Worth, Houston, Laredo, Lufkin,
Paris, Pharr, San Angelo, San Antonio, Wichita Falls, and Yoakum. The responses to the survey
questions so far are summarized in this section.
Question 1: Have you used or are you using Asphalt Treated Base (ATB) in construction/
rehabilitation projects in your district?
Six districts (Abilene, Houston, Paris, Pharr, San Angelo, and Wichita Falls) reported
they had used or were using ATB in their projects. Based on information from lettings in Site
Manager, the districts that had recently used Item 292 are as per Table 3.1. The districts that
have used Item 345 are summarized in Table 3.2.
Out of the responses obtained, six districts (Abilene, Houston, Paris, Pharr, San Angelo,
and Wichita Falls) reported they have used or are using ATB in their projects.
Table 3.1: Districts that have used ATB in their construction based on Item 292.
Tons of Asphalt Treated Base Material
Year Abilene Beaumont Houston Paris
San
Angelo
San
Antonio Waco
Wichita
Falls
2005 2,882 16,602
2006 139,815 149,159 248 217 688
2007 106,890 87,341 49,530 1,222 318
2008 9,273 59,872 430,632 13,430 3,123
30
Table 3.2: Districts that have used ATB in their construction based on Item 345.
Tons of Asphalt Treated Base Material
Year Beaumont El Paso Houston San Angelo San Antonio Waco
2005 251,204 74,751 13,359 1,394 16,616
2006 122,402 220,980 133
2007 162,887 441 146,029 3,103
2008 43,679 353,711
Based on information from lettings in Site Manager, the districts that have recently used
Item 292 are as per Table 3.1. The districts that have used Item 345 (similar but discontinued
version of Item 292) are summarized in Table 3.2.
Houston and Beaumont are the districts with the highest quantities of ATB. Personal
contact with the Wichita Falls and San Antonio Districts indicated that they had changed ordered
the ATBs reflected in the tables to Type A/B HMA mixes.
Question 2: If yes, how many such projects have been completed in the last 5 years or are
scheduled to be constructed in the near future in your district?
The six districts that responded positively to the previous question have at least worked
with one ATB project, with Houston being the leader with approximately 20 ATB projects.
Question 3: Which specification do you use for the design of ATB?
Of the districts that use ATB, 50% follow only Item 292 for the design of ATB. Abilene
District uses only Tex-204-F for the design, and Pharr district applies both design procedures
(Figure 3.1).
If you use Item 292 or Tex-204-Part III, do you waive any of the requirements?
Abilene waives Tex 242-F “Hamburg Wheel-Tracking Test”, Tex 226-F “Indirect
Tensile Strength”, and if virgin base is used, it should meet triaxial requirements as per Tex 126-
F “Molding, Testing, and Evaluating Bituminous Black Base Material”. Houston and Pharr
waive Tex-126-E and the strength requirements, respectively.
31
50%
17%
33%
17%
Item 292
Tex 204-F part III
Both
Other
Percentage of District
Figure 3.1: Specifications Used for ATB Design.
Question 4: Which compactor do you use for the design of ATB?
As shown in Figure 3.2, half of the districts that handle ATB use the Superpave Gyratory
Compactor (SGC) for their design. Only San Angelo district uses the 6 in. gyratory press
mentioned in Tex-126-E.
33%
50%
0%
17%
TGC
SGC
Both
Other
Percentage of District
Figure 3.2: Compactors Used for ATB Design.
32
Question 5: What are the main uses of ATB in your district?
Most of the districts make use of ATB as an alternative to stabilized base and to Type
A/B hot mix asphalt (HMA). Pharr District also applies ATB to reduce the pavement structure
by eliminating the lime-treated subgrade at high volume intersection.
Question 6: What factors motivate you to select ATB for projects in your district over
other alternatives?
The respondents indicated their main reasons of using ATB are the following items: (1)
more economical, (2) easier to construct, (3) short curing time, (4) stronger than stabilized bases,
and (5) rut resistance (see Figure 3.4).
Question 7: What typical aggregate types does your district use on ATB projects?
Based on the responses to the questionnaire, the majority of the districts use limestone as
aggregate for ATB, just Paris district uses sandstone as aggregate.
33%
67%
67%
0%
17%
Alternative to unbound base
Alternative to stabilized base
Alternative to Type A/B HMA
Alternative to bond breaker under PCC
Other
Percentage of District
Figure 3.3: Main Uses of ATB.
33
50%
0%
33%
0%
50%
More economical
Lack of aggregates for alternatives
Easier to construct
Easier to incorporate recycled materials
Other
Percentage of District
Figure 3.4: Factors for Selection of ATB in Projects.
Question 8: Do you add RAP or Crushed Concrete to your ATB?
50% of the districts using ATB add RAP to ATB, the other half do not add RAP or
crushed concrete.
Question 9: As per Item 292, what are the major types and grades of the materials you use
in your district?
Grade 1 as per Item 292 is the most used throughout the state, but Abilene and Houston
Districts prefer to use Grade 2.
Question 10: What binder grades does your district use on ATB projects?
As shown in Figure 3.5, PG 64-22 is used for ATB by all districts in Texas. However,
occasionally PG 70-22 and PG 76-22 are specified.
Question 11: What criteria are used to determine the amounts of binder?
Most of the districts that responded positively to the use of ATB in projects determine the
amount of binder following TxDOT specifications, mainly density. San Angelo district bases the
amount of binder on Area Engineer’s preference and experience.
34
Question 12: What construction specifications do you use for your projects?
Half of the districts follow Item 292 for construction purposes. Pharr District states a lift
thickness no greater than 4 in. Paris District requires 5% to 9% air voids calculated by the
Theoretical Maximum Specific Gravity.
100%
33%
17%
PG 64-22
PG 70-22
PG 76-22
Percentage of District
Figure 3.5: Binder Used for ATB Projects.
Question 13: What types of problems, if any, have you encountered with design or
construction of ATB?
Based on the questionnaire, the districts were satisfied with the performance of their
projects. However, segregation is mentioned as a more frequent problem presented in ATB than
in HMA.
The Houston and Paris Districts staff was visited to obtain insight in their use of ATB,
their current mix design processes and their concerns. The insight gained by the research staff
from these two districts was quite evaluable.
Based on the questionnaire and interaction with the PMC, the candidate districts that
were considered for this study are shown in Table 3.3.
35
Table 3.3: Candidate Districts Considered for This Study.
District Aggregate Type
El Paso Dolomite
Beaumont Limestone
Paris Sandstone
Wichita Falls Limestone
Houston LP-610 Limestone
Houston SH-99 Limestone
36
Chapter Four: Comprehensive Evaluation of Alternative Protocols
The current methods commonly used by the districts were described in Chapter 2. This
section contain a deep explanation of those methods along with two alternative proposed
methods that are more in line with the operational requirements of TxDOT. This chapter
addresses the Index properties; OAC based on many parameters; density, strength and modulus
comparisons for prepared specimens following the current and proposed protocols.
Table 4.1 shows a summary of the different mix design methods used in this study. The
main differences between mix design methods are; how the OAC is calculated, the size of the
specimens, curing temperature and the strength test. To set the criteria for each method and to
select a preferred method, a comparative study of the properties of different mixes was carried
out.
Table 4.1: Summary of Mix Design Methods.
Baseline Mix Design Tex-126-E Tex-204-F Tex-126-H Tex-204-H
Compactor TGC SGC SGC SGC
Specimen Size, in. 6x8 6x4.5 6x8 6x4.5
OAC based on Asphalt Density
Curve
Volumetric
Properties
Asphalt
Density
Curve
Asphalt
Density
Curve
Curing and Testing
Temperature, °F 140 77 140 77
Strength Test UCS IDT UCS IDT
4.1 TEX-126-E
In this method, asphalt stabilized (black base) materials are molded using a Texas
Gyratory Compactor (TGC). Several specimens with different asphalt contents are prepared and
tested to develop an asphalt content-density curve and an asphalt content-unconfined
37
compressive strength (UCS) curve. The optimum asphalt content (OAC) is determined from the
asphalt content-density curve. The compacted black base specimens are made in duplicates and
are tested for their unconfined compressive strengths at 140°F in this test method. The
specimens are subjected to two types of deformation rates: slow (0.15 in./min) and fast (10
in./min.) Specimens tested at a slow deformation rate yield relatively lower UCS’s as compared
to the fast deformation rate ones. From the UCS-density relationship, the minimum density that
would satisfy the UCS requirements as per Item 292 is taken as the minimum allowable density.
4.2 TEX-204-F
This procedure was originally developed to design Type A and Type B hot mix at 96%
density using 6 in. x 4½ in. specimens. Unless otherwise indicated, the specimens are molded to
100 gyrations using an SGC. The quality of the mix assessed with the indirect tensile strength at
a deformation rate of 2 in./min and a nominal temperature of 77oF.
4.3 PROPOSED TEX-126-H
The proposed Tex-126-H (hybrid) is similar to Tex-126-E, with one exception. While
Tex-126-E uses a Texas Gyratory Compactor (TGC), Tex-126-H advocates the use of a
Superpave Gyratory Compactor (SGC) to produce the specimens. The rest of the protocol is the
same as Tex-126-E, where the asphalt-density curve and UCS-density curve are developed to
assess the OAC and field density.
4.4 PROPOSED TEX-204-H
The second proposed method, Tex-204-H (Hybrid), is similar to Tex-204-F. The only
difference is the way the OAC is selected. For Tex-204-H, the requirements for volumetric
properties such as VFA are waived because it is not possible to modify the gradation to achieve
the required volumetric properties. In Tex-204-H protocol, attention is paid to asphalt-density
curve, similar to Tex-126-H, and the IDT strength.
38
4.5 COMPARISON OF PROPERTIES FROM DIFFERENT APPROACHES
Mix designs were performed on six materials following the four approaches indicated
above to highlight their similarities and differences. These six materials were collected from
different sites from El Paso, Beaumont, Paris, Wichita Falls and two sites from Houston, LP-610
and SH-99. To demonstrate different approaches a local material from El Paso is used as an
example. Results from the other materials are then summarized.
4.5.1 Index Properties
The gradation curves of all materials as received are shown in Figure 4.1. To prepare the
materials for mix design, the entire stock of the material was sieved first to develop a global
gradation curve. This gradation curve was used throughout the study. The acceptable range as
per Item 292 Grade 1 is also shown in the figure. Houston SH-99 and Paris materials do not
fulfill those requirements.
0
10
20
30
40
50
60
70
80
90
100
0.010.1110100Sieve size, mm
Perc
ent P
assin
g
El PasoBeaumontHouston SH-99Houston I-60ParisWichita FallsItem 292
#4 #40 #200
Gravel Sand Fines
LP-610
0
10
20
30
40
50
60
70
80
90
100
0.010.1110100Sieve size, mm
Perc
ent P
assin
g
El PasoBeaumontHouston SH-99Houston I-60ParisWichita FallsItem 292
#4 #40 #200
Gravel Sand Fines
0
10
20
30
40
50
60
70
80
90
100
0.010.1110100Sieve size, mm
Perc
ent P
assin
g
El PasoBeaumontHouston SH-99Houston I-60ParisWichita FallsItem 292
#4 #40 #200
Gravel Sand Fines
LP-610
Figure 4.1: Gradation Curves from Different Materials Used in This Study.
39
Figure 4.2 illustrates the material constituents for all materials. Houston SH-99 and Paris
have higher concentrations of gravel (72% and 70%, respectively) while El Paso material has the
lowest gravel concentration (55%). Coarse sand contents range from 9% to 28%, with Houston
SH-99 containing the lowest concentration and Wichita Falls the highest. The fine sand contents
of all materials are similar, ranging from 12% to 18%. El Paso and Wichita Falls have the
highest fines concentrations (5%) while other materials contain 1% to 2% fines.
Figure 4.2: Material Constituents for the Sites Being Studied.
Material classifications as per Unified Soil Classification System (USCS) and American
Association of State Highway and Transportation Officials (AASHTO) Soil Classification
System, as well as the Atterberg limits are summarized in Table 4.2. Under the USCS, all
materials classified as GP (poorly-graded gravel) except for Houston LP-610 and Wichita Falls
which were categorized as GW (well-graded gravel). All materials are classified as A-1-a under
the AASHTO system. All materials seem to be non-plastic except Wichita Falls.
40
Table 4.2: Soil Classification and Plasticity Index for Bases under Study.
Classification Atterberg Limits Material Source
USCS AASHTO Liquid Limit Plasticity Index
Beaumont GP A-1-a Non Plastic
El Paso GP A-1-a Non Plastic
Houston SH-99 GP A-1-a Non Plastic
Houston LP-610 GW A-1-a Non Plastic
Paris GP A-1-a Non Plastic
Wichita Falls GW A-1-a 21 4
The sand equivalency, wet ball mill and aggregates hardness for each material are
presented in Table 4.3. The materials used in this study met the sand equivalency requirement
except for Wichita Falls materials. All material met the wet ball mill requirements; Beaumont
material was near to reach the maximum acceptable value. All materials exhibit Aggregate
Crushing Value (ACV) and Aggregate Impact Value (AIV) 30% or less.
Table 4.3: Sand Equivalency and Wet Ball Mill and Hardness for Bases under Study.
Material Source Sand
Equivalency, % Wet Ball Mill, %
Aggregate
Crushing Value
Aggregate
Impact Value
Beaumont 79 49 15 8
El Paso 53 28 16 9
Houston SH-99 82 11 25 16
Houston LP-610 80 16 29 18
Paris 52 12 24 14
Wichita Falls 26 31 30 20
Item 292 Limits Min 40 Max 50 - -
41
4.5.2 Results from Tex-126-E Protocol
Three sets of 6 in. by 8 in. specimens, prepared with the TGC at nominal asphalt contents
of 3%, 4.5% and 6%, were tested to determine the optimum asphalt content (OAC). Based on
the consultation with the PMC, the maximum AC content was limited to 6% for economic
reasons. A minimum AC content of 3% was selected for constructability.
The compacted specimens were weighed and their heights and diameters were measured
to calculate their densities. After the specimens equilibrated to 140°F oven for 48 hrs, they were
tested as quickly as possible to avoid any heat loss for their moduli using the Free-Free Resonant
Column (FFRC) tests, followed by UCS. Figure 4.3 shows the results from the El Paso material.
The OAC as per Tex-126-E is about 5.6%. The strength requirements for ATB as per
Item 292 (50 psi) are met by all three AC contents used in this study.
Two alternative OACs can also be extracted from Figure 4.3, one being at the AC content
when the modulus is maximum and the other when the UCS is maximum. The OACs based on
modulus and UCS are 4.1% and 3.0%, respectively.
Table 4.4 shows the density, UCS and modulus for each sets of specimens tested at 3%,
4.5% and 6% asphalt contents for all materials. The variations in density with AC content are
6% or less for all materials. Almost all materials exhibited their lowest strengths and moduli at
an AC content of 6%.
The OAC values based on density are compared to those from modulus and strength in
Table 4.5 for all materials. The OACs based on density are typically higher as compared to
those based on modulus and strength. The OACs based on densities fall between 4.3% and 6%.
42
0
200
400
600
800
1000
2 3 4 5 6 7Asphalt Content, %
Mod
ulus
, ksi
130
135
140
145
150
Density, pcf
Modulus Density
a) Asphalt Content–Density/Modulus Curves
0
50
100
150
200
250
300
2 3 4 5 6 7Asphalt Content, %
Unc
onfin
ed C
ompr
essi
veSt
reng
th, p
si
130
135
140
145
150
Density, pcf
Unconfined Compressive Strength Density
b) Asphalt Content-Density/Strength Curves
Figure 4.3: Density/Modulus /Strength vs. Asphalt Content as per Tex-126-E.
4.5.3 Results from Tex-126-H Protocol
The same process followed to obtain the OACs using Tex-126-H specifications. A Pine
SGC at 100 gyrations was used to prepare the specimens. Figure 4.4 shows the asphalt content-
density, asphalt content-modulus and asphalt content-strength curves for El Paso material as per
43
Table 4.4: Density, Strength and Modulus at Different Asphalt Contents as per Tex-126-E.
Material Site
Parameter Asphalt
Content Beaumont El Paso Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
3% 137 138 138 140 129 128
4.5% 140 143 144 141 129 132 Density, pcf
6% 141 145 141 140 130 136
3% 161 192 206 76 214 227
4.5% 126 169 95 160 141 354 UCS, psi
6% 119 55 69 85 128 236
3% 609 526 865 560 650 653
4.5% 558 871 690 1145 687 1443 Modulus, ksi
6% 271 436 481 623 426 962
Table 4.5: Optimum Asphalt Content based on Density, Strength or Modulus as per Tex-126-E.
Optimum Asphalt
Content Based on
Maximum
Beaumont El Paso Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
Density 6.0 5.6 4.7 4.5 4.3 6.0
UCS 6.0 3.0 3.0 4.5 3.0 4.5
Modulus 3.4 4.1 3.0 4.7 4.0 4.8
Tex-126-H. The specimens with 3% binder exhibited the highest modulus and strength. OAC
from maximum density was obtained at 3.9% asphalt content. The density curve in Figure 4.4 is
so flat that any asphalt content between 3% and 5% can easily be practically considered as the
OAC from that curve given the uncertainties in obtaining the density of each specimen.
44
0
200
400
600
800
1000
2 3 4 5 6 7Asphalt Content, %
Mod
ulus
, ksi
130
135
140
145
150
Density, pcf
Modulus Density
a) Asphalt Content–Density/Modulus Curves
0
50
100
150
200
250
300
2 3 4 5 6 7
Asphalt Content, %
Unc
onfin
ed C
ompr
essi
veSt
reng
th, p
si
130
135
140
145
150
Density, pcf
Unconfined Compressive Strength Density
b) Asphalt Content-Density/Strength Curves
Figure 4.4: Density/Modulus /Strength vs. Asphalt Content as per Tex-126-H.
The density, UCS and modulus of every sets of specimens tested are summarized in
Table 4.6. The variations in the densities with AC contents were less than 5% for all materials.
The densities are generally the greatest at binder contents of 4.5%. Ignoring Wichita Falls
45
Table 4.6: Density Strength and Modulus at Different Asphalt Contents as per Tex-126-H.
* unable to retrieve an intact specimen after compaction and as such could not be tested.
materials, the maximum strengths and moduli were obtained at 3% AC contents. Once again, all
specimens marginally or significantly passed the strength requirement of 50 psi for ATB as per
Item 292. As reflected in Table 4.7, once again the highest OACs are generally obtained based
on density.
Table 4.7: Optimum Asphalt Contents Based on Density, Strength or Modulus as per Tex-126-H.
Optimum Asphalt
Content Based on
Maximum
Beaumont El Paso Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
Density 4.7 3.9 4.2 4.1 6.0 6.0
UCS 4.1 3.0 3.0 4.2 5.0 4.7
Modulus 3.6 3.0 3.4 3.0 5.5 5.4
Parameter Asphalt
Content Beaumont El Paso
Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
3% 134 141 135 136 127 126
4.5% 136 141 136 137 126 130 Density, pcf
6% 135 138 134 134 130 133
3% 159 230 84 167 N/A* 207
4.5% 180 192 55 121 163 410 UCS, psi
6% 105 73 48 70 125 301
3% 948 664 1022 1422 N/A* 498
4.5% 960 537 915 909 527 1423 Modulus, ksi
6% 462 115 369 534 600 1524
46
4.5.4 Results from Tex-204-H Protocol
The asphalt-density curve for Tex-204-H for El Paso material is illustrated in Figure 4.5.
The OAC based on density is 3.9% at a density of about 139 pcf. It is hard to judge the accuracy
of this OAC, since the density changes by less than 1% with the change in the asphalt content.
The OACs based on modulus and IDT occur at about 5.3% and 3.7%, respectively. Since the
specimens’ heights were only 4.5 in., a v-meter was used to measure the moduli.
0
200
400
600
800
1000
1200
2 3 4 5 6 7
Asphalt Content, %
Mod
ulus
, ksi
138
139
140
Density, pcf
Modulus Dry Density
a) Asphalt Content-Density/Modulus Curves
0204060
80100120140
2 3 4 5 6 7
Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h,
psi
138
139
140
Density, pcf
Indirect Tensile Strength Density
b) Asphalt Content-Density/Strength Curves
Figure 4.5: Density/Modulus /Strength vs. Asphalt Content as per Tex-204-H.
47
The values obtained for density, indirect tensile strength (IDT) and modulus for all
materials prepared under Tex-204-H are summarized in Table 4.8. For all materials the
variations in the density with asphalt content are rather small. Most materials exhibit their
highest IDT strengths between asphalt contents of 3 and 4.5%. Wichita Falls specimens with 3%
asphalt content could not be tested because they broke after compaction but before testing.
Moduli of the specimens tested do not vary much for most materials similar to the densities.
The OAC values based on maximum density, indirect tensile strength and modulus as per
Tex-204-H for all materials are summarized in Table 4.9. As opposed to the Tex-126-E and
Tex-126-H results, no clear pattern is apparent for this protocol.
Table 4.8: Density, Strength and Modulus at Different Asphalt Contents as per Tex-204-H.
Parameter Asphalt
Content Beaumont El Paso
Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
3% 133 139 138 137 126 129
4.5% 136 139 138 138 130 132 Density, pcf
6% 135 139 135 134 130 134
3% 133 124 118 168 91 N/A
4.5% 163 121 105 150 116 110 IDT
Strength, psi 6% 100 47 72 104 101 170
3% 855 783 1007 538 617 N/A
4.5% 862 1030 1103 553 622 790 Modulus, ksi
6% 880 1030 1055 547 688 857
48
Table 4.9: Optimum Asphalt Contents Based on Density, Strength or Modulus as per Tex-204-H.
Optimum Asphalt
Content Based on
Maximum
Beaumont El Paso Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
Density 6.0 3.9 3.8 3.8 6.0 6.0
IDT Strength 4.2 3.7 4.7 3.0 6.0 6.0
Modulus 6.0 5.3 4.8 4.8 6.0 5.6
4.1.5 Tex-204-F
Figure 4.6 shows the gradation curves for the six materials and the acceptable limits for
Item 344 type SP-A. El Paso, Houston LP-610 and Wichita Falls materials fit well between the
gradation requirements for Item 344 Grade SP-A. Beaumont, Houston SH-99 and Paris
materials do not fulfill those requirements, since they are coarser than Item 344 Grade SP-A
requirements.
0
10
20
30
40
50
60
70
80
90
100
0.010.1110100Sieve size, mm
Perc
ent P
assi
ng
El PasoBeaumontHouston SH-99Houston I-60ParisWichita Falls344 SP-A
#4 #40 #200
Gravel Sand Fines
Figure 4.6: Gradation Curves from Different Materials Compared to Item 344 Grade SP-A.
49
To check whether any of the mixes would meet the volumetric properties as per Tex-204-
F Part III, the theoretical maximum specific gravity (Gmm) in accordance to Tex-227-F were
determined. Triplicate samples were tested to assess the repeatability of this test method. The
Gmm values are reported in Table 4.10. All samples tested exhibited coefficients of variation
(COV’s) of 1.2% or less.
The bulk specific gravities (Gmb) for all specimens are presented in Table 4.11. The
maximum Gmb values are obtained at binder contents between 4.5 and 6%.
Table 4.10: Maximum Theoretical Specific Gravities for Tex-204-H Specimens.
Asphalt
Content Parameter Beaumont El Paso
Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
Average 2.45 2.62 2.51 2.47 2.37 2.55 3.0%
COV* 0.3% 0.5% 0.3% 0.4% 0.1% 0.4%
Average 2.45 2.55 2.43 2.41 2.46 2.48 4.5%
COV 0.2% 0.7% 0.1% 0.1% 0.3% 0.1%
Average 2.39 2.46 2.38 2.36 2.4 2.41 6.0%
COV 0.3% 1.2% 0.4% 0.2% 0.7% 1.1% * COV = coefficient of variation
Table 4.11: Bulk Specific Gravities for Tex-204-H Specimens.
Asphalt
Content Beaumont El Paso
Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
3.0% 2.33 2.41 2.39 2.30 2.25 2.28
4.5% 2.37 2.47 2.40 2.35 2.30 2.37
6.0% 2.41 2.40 2.41 2.30 2.34 2.41
50
Figure 4.7 shows an example of the volumetric properties for the El Paso material. For
El Paso material, the OAC is 5.2% (see Figure 4.7a). Voids in mineral aggregates (VMA) for
5.2% asphalt content is about 15%, which is greater than a minimum value of 13 specified in
Item 344 for SP-A mixes. The voids filled with asphalt (VFA) of about 74% at a binder content
of 5.2% fell between the 65-75% as required by Item 344.
y = -2.0x + 14.4
R2 = 0.9
0123456789
10
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5Asphalt Content, %
Air
Voi
ds, %
a
a)
y = -2.0x + 14.4
R2 = 0.9
0123456789
10
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5Asphalt Content, %
Air
Voi
ds, %
a y = -2.0x + 14.4
R2 = 0.9
0123456789
10
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5Asphalt Content, %
Air
Voi
ds, %
a
a)y = 0.9x2 - 6.9x + 27.6
R2 = 1.0101112131415161718
2.5 3.5 4.5 5.5 6.5Asphalt Content, %
VM
A, %
b
b) y = 0.9x2 - 6.9x + 27.6R2 = 1.0
101112131415161718
2.5 3.5 4.5 5.5 6.5Asphalt Content, %
VM
A, %
b
y = 0.9x2 - 6.9x + 27.6R2 = 1.0
101112131415161718
2.5 3.5 4.5 5.5 6.5Asphalt Content, %
VM
A, %
b
b)
y = 14.6x - 1.9R2 = 1.0
3040
5060
7080
90
2.5 3.5 4.5 5.5 6.5Asphalt Content, %
VFA
, %
c
c)y = 14.6x - 1.9
R2 = 1.03040
5060
7080
90
2.5 3.5 4.5 5.5 6.5Asphalt Content, %
VFA
, %
c
y = 14.6x - 1.9R2 = 1.0
3040
5060
7080
90
2.5 3.5 4.5 5.5 6.5Asphalt Content, %
VFA
, %
c
c)
Figure 4.7: Volumetric Properties for El Paso Material.
Table 4.12 shows the volumetric properties for all specimens as well as the requirements
to be met for Item 344 SP-A mixes. The majority of the materials successfully passed the
requirements, except for the VFA value for Houston SH-99, Paris and Wichita Falls.
51
Table 4.12: Volumetric Properties for Tex-204-H Specimens.
Values at 4%
Air Voids Beaumont El Paso
Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
Item 344
Grade
SP-A
AC, % 4.0 5.2 3.7 4.3 5.8 5.0 -
VMA, % 14.3 14.9 20.2 14.8 18.4 15.6 Min. 13
VFA, % 71.5 74.1 80.4 72.4 78 75.5 65-75
4.5.6 Moisture Susceptibility
One of the test methods used to quantify moisture susceptibility is Tex-242-F, Hamburg
Wheel-Tracking Test (HWTD). This test was performed on specimens prepared at OAC’s
determined from Tex-126-H and Tex-204-H protocols based on density. The results for the
HWTD are summarized in Table 4.13.
Table 4.13: Hamburg Wheel Test Results for All Materials.
Rut Depth, in Test
Method
Number of
Passes Beaumont El PasoHouston
SH-99
Houston
LP-610 Paris
Wichita
Falls
OAC, % 4.7 3.9 4.2 4.1 6.0 6.0
10,000 0.089 0.17 0.207 0.219 0.088 0.108
15,000 0.096 0.211 0.312 0.45 0.092 0.115
Tex-126-
H
20,000 0.124 0.235 0.463 0.478 0.111 0.135
OAC, % 6.0 3.9 3.8 3.8 6.0 6.0
10,000 0.009 0.151 0.102 0.263 0.088 0.108
15,000 0.016 0.174 0.108 0.482 0.092 0.115
Tex-204-
H
20,000 0.023 0.185 0.14 * 0.111 0.135 * Exceeded 0.5 in. rut after 15,300 passes.
52
were continued up to 20,000 passes. All specimens tested exhibited acceptable rut depths at
10,000 cycles. All specimens were in the acceptable limits at 15,000 and 20,000 cycles except
for Houston LP-610 which reached the 0.5 in. deformation at 15,300 cycles.
The second test used to quantify moisture susceptibility is Tex-144-E, the Tube Suction
Test (TST). This test was also performed using the OAC based on density obtained in this study
for Tex-126-H and Tex-204-H. Table 4.14 shows the strengths and moduli before and after the
TST tests. The difference between before and after the TST is the asphalt curing time. Most of
the materials demonstrated an increase in strength and moduli after the TST. Houston LP-610
was the only material that did not gain strength and modulus with curing.
4.5.7 Analysis of Results
The OAC values based on the three different criteria (density, strength and modulus) used for all
materials are summarized in Figure 4.8. The variations in OACs based on density for Tex-126-
H, Tex-204-H and Tex-204-F are compared with the OACs based on density for Tex-126-E in
Figure 4.9. The alternative mix design methods typically yield lower OACs as compared to Tex-
126-E. Paris material was the only one that had higher OAC’s with Tex-126-H and Tex-204-H
than OAC for Tex-126-E. This inconsistency is caused by the flatness of the density curves as
discussed above. In general, the density is not very sensitive to asphalt content as discussed
above. As such, the OAC obtained based on density should be reviewed to decide on the OAC
of a mix.
Similarly, the OACs based on strength are compared in Figure 4.10. Most of the OACs
are greater than those of the Tex-126-E OAC values based on strength. Two of the materials,
Beaumont and Houston LP-610 were outliers; they had a lower OAC than Tex-126-E in all three
different mix design methods.
The OAC values based on modulus are compared in Figure 4.11. Similarly most of the
OACs are greater than that of Tex-126-E values based on modulus with three outliers Houston
LP-610 (Tex-126-H and Tex-204-F) and El Paso (Tex-126-H).
53
Table 4.14: Impact of Tube Suction Tests on UCS and Seismic Modulus.
Beaumont El Paso Houston SH-
99
Houston LP-
610 Paris Wichita Falls Test
Method Parameter
Before After Before After Before After Before After Before After Before After
OAC, % 4.7 3.9 4.2 4.1 6.0 6.0
UCS, psi 405 509 485 653 373 498 583 539 337 446 392 426
Tex
-126
-H
Modulus,
ksi 2411 2242 1739 3028 1279 2738 2494 2462 1720 2482 1940 2054
OAC, % 6.0 3.9 3.8 3.8 6.0 6.0
IDTS, psi 92 107 123 190 63 148 203 153 98 134 143 212
Tex
-204
-H
Modulus,
ksi 833 938 825 901 835 1123 910 834 872 919 882 834
54
6.0
5.6
4.7
4.5
4.3
6.0
4.7
3.9 4.
2
4.1
6.0
6.0
6.0
3.9
3.8
3.8
6.0
6.0
4.0
5.2
3.7
4.3
5.0
0
1
2
3
4
5
6
7
Beaumont El Paso HoustonSH-99
HoustonLP-610
Paris WichitaFalls
Opt
imum
Asp
halt
Con
tent
, %
Tex-126-E Tex-126-H Tex-204-H Tex-204-F
a) Density
6.0
3.0
3.0
4.5
3.0
4.5
4.1
3.0
3.0
4.2
5.0
4.7
4.2
3.7
4.7
3.0
6.0
6.0
0
1
2
3
4
5
6
7
Beaumont El Paso HoustonSH-99
HoustonLP-610
Paris WichitaFalls
Opt
imum
Asp
halt
Con
tent
, %
.
Tex-126-E Tex-126-H Tex-204-H
b) Strength
3.4
4.1
3.0
4.7
4.0
4.8
3.6
3.0 3.
4
3.0
5.5
5.4
6.0
5.3
4.8
4.8
6.0
5.6
0
1
2
3
4
5
6
7
Beaumont El Paso HoustonSH-99
HoustonLP-610
Paris WichitaFalls
Opt
imum
Asp
ahal
t Con
tent
, %
.
Tex-126-E Tex-126-H Tex-204-H
c) Modulus
Figure 4.8: Comparison of Optimum Asphalt Contents Based on Density, Strength, and Modulus.
55
0
1
2
3
4
5
6
7
0 1 2 3 4 5 6 7Tex-126-E OAC, %
OA
C B
ased
on
othe
r Pr
otoc
ols,
%Tex-126-H
Tex-204-H
Tex-204-F(Interpolated from Tex-204-H)
Figure 4.9: Comparison of Tex-126-E OAC with other Methods Based on Density.
0
1
2
3
4
5
6
7
0 1 2 3 4 5 6 7Tex-126-E OAC, %
OA
C B
ased
on
othe
r Pr
otoc
als,
% Tex-126-H
Tex-204-H
Tex-204-F(Interpolated from Tex-204-H)
Figure 4.10: Comparison of Tex-126-E OAC with other Methods Based on Strength.
56
0
1
2
3
4
5
6
7
0 1 2 3 4 5 6 7Tex-126-E OAC, %
OA
C B
ased
on
othe
r Pr
otoc
ols,
%Tex-126-H
Tex-204-H
Tex-204-F(Interpolated from Tex-204-H)
Figure 4.11: Comparison of Tex-126-E OAC with other Methods Based on Modulus.
Figure 4.12 shows the maximum densities for all materials and for all mix design
methods. For a given mix, the densities are fairly close. The maximum density for most
materials is generally reached when Tex-126-E is followed. The lowest densities are typically
obtained from Tex-204-F by waiving some of the volumetric requirements.
For comparison purposes, UCS and IDT specimens were prepared based on OAC of each
method and were tested for strength and modulus. The variations in the UCS values are shown
in Figure 4.13. All materials reached the 50 psi strength requirement for ATB as per Item 292.
The strongest material is Wichita Falls while the weakest material analyzed was Houston SH-99.
The specimens prepared with the Superpave Gyratory Compactor usually exhibit similar or
higher UCS as compared to those prepared with the Texas Gyratory Compactor.
The IDT strengths obtained from 6 x 4.5 in. specimens are shown in Figure 4.14. No
obvious pattern could be found in the data. Given the uncertainty in the IDT tests, in most cases
the strengths are similar from different methods of estimating OAC.
57
141
145
144
141
129
13613
7
141
136 13
8
129
133
137 13
9
138 13
9
130
13413
6
139
136 13
8
133
120
125
130
135
140
145
150
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Den
sity,
pcf
Tex-126-E Tex-126-H Tex-204-H Tex-204-F*
Figure 4.12: Densities at OACs for Each Test Method.
124
92 86
160
147
236
174
219
60
169
125
301
105
219
65
142
125
301
184
149
68
126
408
0
50
100
150
200
250
300
350
400
450
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Unc
onfin
ed C
ompr
essiv
e St
reng
th, p
si .
Tex-126-E Tex-126-H Tex-204-H* Tex-204-F*
Figure 4.13: UCS at OACs for Each Test Method.
58
100
73
100
150
115
170
161
130
113
156
101
170
100
130
114
169
101
170
163
97
158
153
135
0
20
40
60
80
100
120
140
160
180
200
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Indi
rect
Ten
sile
Stre
ngth
, psi
Tex-126-E* Tex-126-H* Tex-204-H Tex-204-F
Figure 4.14: IDT at OACs for Each Test Method.
Finally, the moduli obtained from these specimens are shown in Figures 4.15a and 4.15b.
The variations in moduli are in line with the corresponding strengths shown in Figures 4.13 and
4.14. So far this study shows that mix design using the SGC is feasible. Based on the results
reported in this chapter, the adoption of either Tex 126-H or Tex 204-H will be discussed.
59
272
601 66
1
1137
695
569
874
615
964 10
32
600
1524
462 61
5
1006 11
43
600
1524
956
385
1012
974
1582
0
200
400
600
800
1000
1200
1400
1600
1800
2000
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Seism
ic M
odul
us, k
si
Tex-126-E Tex-126-H Tex-204-H* Tex-204-F*
a) FFRC Moduli on 6 in. by 8 in. Specimens
880
1066 11
07
522 61
9
857
864 95
0 1069
550 61
1
857
880 95
0 1064
1032
611
857
858
1073
857
552
855
0
200
400
600
800
1000
1200
1400
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Seism
ic M
odul
us, k
si
Tex-126-E* Tex-126-H* Tex-204-H Tex-204-F
b) V-meter Moduli on 6 in. by 4.5 in. Specimens
Figure 4.15: Seismic Modulus at OACs for Each Test Method.
60
Chapter Five: Evaluation of Parameters that Impact Performance
The two alternative protocols deemed reasonable for fulfilling the objectives of this
project are Tex-126-H and Tex-204-H. Many specimens were mixed, compacted and tested at
different numbers of gyrations using the SGC to evaluate the impact of the number of gyrations
on the properties of the mixes. Specimens were also mixed and compacted using the TGC to try
to match the densities of the SGC specimens with the ones compacted using the TGC. The
impact of curing temperature is also presented in this chapter in which specimens were tested
after 24 or 48 hrs of curing at 77°F or 140°F. Two different grades of asphalt (PG 76-22 and PG
64-22) were evaluated as well. The differences in optimum asphalt content, density, strength and
modulus between the two binders are presented. The impact of the change in gradation on
optimum asphalt content, density, strength and modulus is summarized at the end of this chapter.
5.1 IMPACT OF NUMBER OF GYRATIONS
5.1.1 Tex-126-H Protocol
The impact of the number of gyrations on the optimum asphalt content (OAC) based on
density as per Tex-126-H is summarized in Figure 5.1. For Paris and Wichita Falls, the OAC’s
seem to be 6% (upper limit imposed on the asphalt content) or greater. Except for the Houston
SH-99, the increase in the number of gyrations results in typically less than 1% change in the
OAC. The OAC’s as per Tex-126-E are also included in Figure 5.1. The OAC’s for 60 to 80
gyrations are typically closer to the OAC’s from Tex-126-E.
The densities at corresponding OAC’s for different numbers of gyrations of SGC are
compared with those from Tex-126-E in Figure 5.2. The impact of the number of gyrations of
SGC on density is rather small (typically less than 3 pcf). In almost all cases, the highest
densities are associated with Tex-126-E protocol.
The impact of the number of gyrations on UCS and modulus are shown in Figures 5.3
and 5.4, respectively. All specimens were prepared to the nominal dimensions of 6 in. by 8 in.
61
5.6
4.6
6.0
4.5
6.0
6.0
4.9
4.7 4.8
4.3
6.0
6.0
4.7
3.9 4.
2
4.1
6.0
6.0
6.0
5.6
4.7
4.5
4.3
6.0
0
2
4
6
8
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Opt
imum
Asp
halt
Con
tent
, %
60 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
Figure 5.1: Impact of Number of Gyrations on OAC Based on Density for Tex-126 –H Protocol.
135
141
136
136
130 13213
7 141
137
137
127 13
2137 14
1
136 13
8
129 13
3
141 14
5
144
141
129
136
100
110
120
130
140
150
160
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Den
sity
, pcf
60 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
Figure 5.2: Impact of Number of Gyrations on Density for Tex-126-H Protocol.
62
210
204
67
93 100
301
169
208
69
134
133
311
174
219
60
169
125
301
124
92 86
160
147
236
0
50
100
150
200
250
300
350
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Unc
onfin
ed C
ompr
essi
ve S
tren
gth,
psi
60 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
Figure 5.3: Impact of Number of Gyrations on UCS for Tex-126-H Protocol.
817
419 56
7
788
373
1117
720
640 71
0
1130
512
1158
874
615
964 10
32
600
1524
272
601 66
1
1137
695
962
0
200
400
600
800
1000
1200
1400
1600
1800
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Seis
mic
Mod
ulus
, ksi
.
60 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
Figure 5.4: Impact of Number of Gyrations on Modulus for Tex-126-H Protocol.
63
The UCS and modulus values mostly follow the same pattern. As the UCS increases the
modulus also increases. A clear pattern is not evident in these results primarily because of the
complex interaction of the compaction energy with the asphalt content. The other complicating
factor is the so-called locking point during the compaction with the SGC. Locking point is
defined as the first gyration in the first occurrence of three gyrations of the same height (Varvik
and Carpenter, 1998). At the locking point aggregates lock together and additional gyrations
may degrade the aggregates. The locking points for all mixes are presented in Figure 5.5.
Except for Paris materials, the locking points are less than 100 gyrations. This indicates that the
use of 100 gyrations may not be desirable for the ATB mixes.
95
80
88
93
100
65
90
74
85 86
100
59
0
20
40
60
80
100
120
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Loc
king
Poi
nt, G
yrat
ions
Tex-126-H (6x8 in. Specimens) Tex-204-H (6x4.5 in. Specimens)
Figure 5.5: Locking Points for Tex-126-H and Tex-126-E Specimens.
A more systematic way of evaluating the impact of the number of gyrations on density is
shown in Figure 5.6a. The densities from the SGC at different numbers of gyrations are
compared with those from the TGC for all asphalt contents tested (i.e., 3%, 4.5% or 6%).
Overall, the densities obtained with the SGC for the three numbers of gyrations are 2 to 3% less
64
60 gyr.y = 0.97x
80 gyry = 0.97x
100 gyry = 0.98 x
120
125
130
135
140
145
150
120 125 130 135 140 145 150Density from Tex-126-E, pcf
Den
sity
from
Tex
-126
-H, p
cf
60 Gyrations 80 Gyrations 100 Gyrations
a)a) Density at 3, 4.5 and 6% AC
80 gyr.y = 0.97x
60 gyr.y = 0.97x
100 gyr.y = 0.97x
120
125
130
135
140
145
150
120 125 130 135 140 145 150
Density from Tex-126-E, pcf
Den
sity
from
Tex
-126
-H, p
cf
60 Gyrations 80 Gyrations 100 Gyrations
b)b) Density at OAC
Figure 5.6: Comparison of Density from Tex-126-E and Tex-126-H Protocols.
65
than those from the TGC. Naturally, the same trends hold for the maximum density as shown in
Figure 5.6b. The numbers of gyrations passed 60 do not seem to impact the density of the mix
much.
As demonstrated in Chapter 4, the change in density with asphalt content is rather small
for the ATB mixes, which may lead to uncertainty in determining the OAC. Another angle in
recommending the number of gyrations is the sensitivity of the asphalt-content-density
relationships as shown in Table 5.1. The sensitivity is defined as the difference between the
maximum and minimum densities at 3%, 4.5% and 6% asphalt contents (used in defining the
asphalt content density relationships) for a given compaction protocol, divided by the average
density of the three measurements. The greater this value is, the better the asphalt content
density curve is defined. In no case the sensitivity is greater than 7%, indicating that perhaps
judgment should be used in estimating the OAC in all cases. However, the preliminary data
from 80 gyrations seem to be the most promising.
Table 5.1: Sensitivity of Density to Asphalt Content for Tex-126-H Protocol.
Sensitivity, (Max. Density – Min Density)/Avg. Density Compaction
Method Beaumont El Paso Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
SGC 60 Gyrations 1% 2% 3% 2% 6% 5%
SGC 80 Gyrations 3% 4% 3% 4% 5% 5%
SGC 100
Gyrations 1% 2% 1% 2% 7% 5%
TGC 3% 5% 4% 1% 1% 6%
66
5.1.2 Tex-204-H Protocol
The impact of the number of gyrations on OAC, density, indirect tensile strength and
modulus are summarized in Figure 5.7 to Figure 5.10 for Tex-204-H protocol. As reflected in
Figure 5.5, the locking points for this protocol are slightly lower than those for Tex-126-H,
perhaps because of the smaller specimen heights 4.5 in. as opposed to 8 in. for Tex-126-H). The
trends for the OAC’s (Figure 5.7) are similar to those presented for Tex-126-H in Figure 5.1
except for the Beaumont material. The patterns for the densities (Figure 5.8) at the OAC’s are
also similar to those from Tex-126-H.
The number of gyrations does not seem to significantly impact the IDT strengths (Figure
5.9) and moduli (Figure 5.10), and the IDT strengths from the SGC and TGC specimens are
closer to one another than the corresponding UCS values from Tex-126-H and Tex-126-E.
The impact of the number of gyrations on density at respective OAC’s of the mixes is
shown in Figure 5.11. The densities obtained from SGC for the three numbers of gyrations are 2
to 3% less than those from the TGC. In general, the numbers of gyrations above 60 do not
significantly or systematically impact the density. As reflected in Table 5.2, the densities of the
specimens prepared at 60 or 80 gyrations exhibit higher sensitivity to the asphalt content. Based
on this exercise, 60 to 80 gyrations may be more appropriate for preparing specimens.
5.2 IMPACT OF CURING TEMPERATURE
5.2.1 Tex-126-H Protocol
The impacts of curing temperature on the UCS and modulus for specimens prepared as
per Tex-126-H are shown in Figures 5.12 and 5.13, respectively. All specimens were prepared at
their corresponding OAC’s at 100 gyrations. The first series of specimens was cured for 24 hours
at room temperature and tested also at room temperature. The second series of specimens was
cured in an oven at 140°F for 48 hours but tested at room temperature, while the third series of
tests was carried out on specimens cured for 48 hours at 140°F and tested at that temperature.
67
6.0
4.8
6.0
3.9
6.0
6.0
6.0
5.1
6.0
4.0
6.0
6.0
6.0
3.9
3.8
3.8
6.0
6.0
6.0
5.6
4.7
4.5
4.3
6.0
0
2
4
6
8
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Opt
imum
Asp
halt
Con
tent
, %
60 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
Figure 5.7: Impact of Number of Gyrations on OAC for Tex-204-H Protocol.
137 14
1
137
139
127 13
2137 14
1
135 13
9
13113
7 139
138
139
130 13
4
141 14
5
144
141
129
136
130
100
110
120
130
140
150
160
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Den
sity
, pcf
60 Gyrations 80 Gyrations 100 Gyrations Tex-160 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
137 14
1
137
139
127 13
2137 14
1
135 13
9
13113
7 139
138
139
130 13
4
141 14
5
144
141
129
136
130
100
110
120
130
140
150
160
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Den
sity
, pcf
60 Gyrations 80 Gyrations 100 Gyrations Tex-1
137 14
1
137
139
127 13
2137 14
1
135 13
9
13113
7 139
138
139
130 13
4
141 14
5
144
141
129
136
130
100
110
120
130
140
150
160
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Den
sity
, pcf
60 Gyrations 80 Gyrations 100 Gyrations Tex-160 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
Figure 5.8: Impact of Number of Gyrations on Density for Tex-204-H Protocol.
68
103 12
3
20
171
108
186
113 13
3
76
168
119
160
100
130
114
169
101
170
100
73
100
150
115
170
0
50
100
150
200
250
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Indi
rect
Ten
sile
Str
engt
h, p
si
60 Gyrations 80 Gyrations 100 Gyrations Tex-12660 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
103 12
3
20
171
108
186
113 13
3
76
168
119
160
100
130
114
169
101
170
100
73
100
150
115
170
0
50
100
150
200
250
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Indi
rect
Ten
sile
Str
engt
h, p
si
60 Gyrations 80 Gyrations 100 Gyrations Tex-126
103 12
3
20
171
108
186
113 13
3
76
168
119
160
100
130
114
169
101
170
100
73
100
150
115
170
0
50
100
150
200
250
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Indi
rect
Ten
sile
Str
engt
h, p
si
60 Gyrations 80 Gyrations 100 Gyrations Tex-12660 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
Figure 5.9: Impact of Number of Gyrations on IDT Strength for Tex-204-H Protocol.
912
889 923 99
2
758 86
2937
872 97
1
561
751
904
880 95
0 1064
548 61
1
857
880
1066 11
07
522 61
9
857
0
200
400
600
800
1000
1200
1400
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Seis
mic
Mod
ulus
, ksi
.
60 Gyrations 80 Gyrations 100 Gyrations Tex-126-E
Figure 5.10: Impact of Number of Gyrations on Modulus for Tex-204-H Protocol.
69
Table 5.2: Sensitivity of Density to Asphalt Content for Tex-204-H Protocol.
Sensitivity, (Max. Density – Min Density)/Avg. Density Compaction
Method Beaumont El Paso Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
SGC 60
Gyrations 6% 5% 1% 3% 4% 5%
SGC 80
Gyrations 3% 6% 1% 3% 9% 3%
SGC 100
Gyrations 3% 3% 2% 3% 6% 4%
TGC 3% 5% 4% 1% 1% 6%
60 gyr.y = 0.97x
80 gyr.y = 0.97x
100 gyr.y = 0.98x
120
125
130
135
140
145
150
120 125 130 135 140 145 150
Density from Tex-126-E, pcf
Den
sity
from
Tex
-204
-H, p
cf
60 Gyrations 80 Gyrations 100 Gyrations
Figure 5.11: Comparison of Density from Tex-126-E and Tex-204-H Protocols.
70
From the first two sets of data, the 48 hours of curing in the oven does not seem to
significantly impact the strength or modulus of the mixes. As such, instead of curing the
specimens for 48 hrs, one can simply cure them for 24 hrs at room temperature to save time and
complications with handling high-temperature specimens.
As expected, the impact of the temperature at the time of testing is more pronounced
comparing the second and third sets of data. Perhaps the UCS and modulus testing can also be
done at the room temperature to mainstream the testing and to improve the reliability of the
results since the operator is not rushed to test the specimen in order to maintain the temperature
of the specimen.
The strengths from the two alternative curing processes are compared with the standard
ones (i.e. 48 hrs of curing at 140°F and testing at 140°F) in Figure 5.14. The strengths from the
Tex-126-H specimens after subjecting them to standard curing were similar to those of Tex-126-
E, indicating that the compactor used for sample preparation has little impact on the UCS
strength. However, the Tex-126-H specimens tested at 77oF exhibit strengths that are about 3.4
times greater than those tested under Tex-126-E protocol, independent of the number of days (1
or 2 days) that the Tex-126-H specimens were cured.
To further verify this concept, the specimens prepared using Tex-126-E protocol were
subjected to three curing regimes and tested. As shown in Figure 5.15, the trends are similar to
those discussed above. There is more scatter in the data that can perhaps be attributed to the
better quality of the specimens prepared by the SGC. The impact of the temperature is also less
evident, perhaps due to the fact that the TGC specimens are denser requiring more than 4 hrs to
cool to room temperature.
71
372
584
337
584
387
396
405
485
373
583
337 39
2
75
217
98
173
133
213
0
100
200
300
400
500
600
700
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Unc
onfin
ed C
ompr
essi
ve S
tren
gth,
psi
24hrs @77°F 48hrs @140°F Cool 48hrs @140°F
Figure 5.12: Impact of Curing Temperature on UCS for Tex-126-H Protocol.
2139 2200
1172
2596
2213
2010
2411
1739
1279
2495
1721 19
40
444
1064
772
1227
772 95
3
0
500
1000
1500
2000
2500
3000
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Seis
mic
Mod
ulus
, ksi
24hrs @77°F 48hrs @140°F Cool 48hrs @140°F
Figure 5.13: Impact of Curing Temperature on Modulus for Tex-126-H Protocol.
72
24 hrs @ 77°Fy = 3.5x
48hrs @ 140°F - 77°Fy = 3.3x
48hrs @140°Fy = 1.1x
0
100
200
300
400
500
600
700
0 100 200 300 400 500 600 700
UCS as per Tex-126-E (48hrs 140 °F- 140 °F)
UC
S at
oth
er C
urin
g an
d T
estin
g Pr
oces
ses
24hrs @77°F 48hrs @140°F - 77°F 48hrs @140°F
Figure 5.14: Comparison of Strengths from Different Curing Regimes using Tex-126-H Specimens with Those from Tex-126-E.
24 hrs @ 77°Fy = 2.9x
48hrs @ 140°F - 77°Fy = 2.4x
0
100
200
300
400
500
600
700
800
0 100 200 300 400 500 600 700 800
UCS as per Tex-126-E (48hrs 140 °F- 140 °F)
UC
S at
oth
er C
urin
g an
d T
estin
g Pr
oces
ses
24hrs @77°F 48hrs @140°F - 77°F
Figure 5.15: Comparison of Strengths from Different Curing Regimes using Tex-126-E Specimens.
73
5.2.2 Tex-204-H Protocol
The impacts of curing temperature on IDT and modulus for specimens prepared as per
Tex-204-H are shown in Figures 5.16 and 5.17, respectively. Once again, the trends are similar
to those from Tex-126-H. As reflected in Figure 5.18, specimens cured either one or two days at
140oF and tested at 77oF yielded similar results. However, when the specimens were tested at
140oF, the strengths were about 4 times less than those tested at 77°F.
5.3 IMPACT OF BINDER GRADE
5.3.1 Tex-126-H Protocol
The OAC’s for the specimens prepared as per Tex-126-H using PG 64-22 and PG 76-22
binders for all materials are shown in Figure 5.19. The OAC’s are less for the PG 64-22 binder
except for the Wichita Falls material. The Paris mixes were not tested because of the lack of raw
materials.
The impact of the binder grade on the densities of the materials is rather small (less than
2 pcf) for most materials as depicted in Figure 5.20. Wichita Falls is the only material that is
experiencing a significant increase in the density with the PG 76-22 binder. The reason for this
pattern is unknown.
The UCS values for the Tex-126-H specimens mixed with different binder grades at their
respective OAC’s are shown in Figure 5.21. Because of the complex interaction of the AC
content, a clear pattern cannot be observed. The specimens with the 76-22 binder seem to be
stronger in compression for the Houston SH-99, Houston LP-610 and Wichita Falls materials.
The Beaumont material showed no impact of the binder grade, and the El Paso material is the
only material in which the PG 64-22 specimen showed a higher UCS value.
74
138
185
112
157
109
157
92
167
63
203
98
143
22
47
29 37
23
51
0
50
100
150
200
250
300
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Indi
rect
Ten
sile
Str
engt
h, p
si
24hrs @77°F 48hrs @140°F Cool 48hrs @140°F
Figure 5.16: Impact of Curing Temperature on IDT Strength for Tex-204-H Protocol.
965
1801
844 10
38
994
75883
3
1153
835 895
872
882
637
911
719
645 78
7
510
0
500
1000
1500
2000
2500
3000
Beaumont El Paso Houston SH-99
Houston LP-610
Paris Wichita Falls
Seis
mic
Mod
ulus
, ksi
24hrs @77°F 48hrs @140°F Cool 48hrs @140°F
Figure 5.17: Impact of Curing Temperature on Modulus for Tex-204-H Protocol.
75
48hrs @ 140°F - 77°Fy = 1.0x
48hrs @140°Fy = 0.2x
0
50
100
150
200
250
0 50 100 150 200 250IDT as per Tex-204-H (24hrs 77 °F)
IDT
at o
ther
Cur
ing
Proc
esse
s
48hrs @140°F - 77°F 48hrs @140°F
Figure 5.18: Comparison of Strengths from Different Curing Regimes using Tex-204-H Specimens.
4.7
3.9 4.
2
4.1
6.0
6.0
4.7
6.0
4.6 4.7
0
2
4
6
8
Beaumont El Paso Houston SH-99
Houston LP-610
Wichita Falls
Opt
imum
Asp
halt
Con
tent
, %
PG 64-22 PG 76-22
Figure 5.19: Impact of Binder Grade on OAC based on Density for Tex-126-H Protocol.
76
137 141
136
138
133
135 14
2
137
137 14
3
0
50
100
150
200
Beaumont El Paso Houston SH-99
Houston LP-610
Wichita Falls
Den
sity
, pcf
PG 64-22 PG 76-22
Figure 5.20: Impact of Binder Grade on Density for Tex-126-H Protocol.
174
219
60
169
301
175
128
96
195
340
0
50
100
150
200
250
300
350
400
Beaumont El Paso Houston SH-99
Houston LP-610
Wichita Falls
Unc
onfin
ed C
ompr
essi
ve S
tren
gth,
psi
PG 64-22 PG 76-22
Figure 5.21: Impact of Binder Grade on UCS for Tex-126-H Protocol.
77
5.3.2 Tex- 204-H Protocol
The OAC’s based on density obtained from the two binders from the Tex-204-H protocol
are compared in Figure 5.22. In this case, the OAC’s from the two binders are closer to one
another. As shown in Figure 5.23, the densities were not impacted significantly by the binder
grade. They are generally within 3 pcf of one another.
6.0
3.9
3.8
3.8
6.0
6.0
4.1
4.8
4.2
5.3
0
2
4
6
8
Beaumont El Paso Houston SH-99
Houston LP-610
Wichita Falls
Opt
imum
Asp
halt
Con
tent
, %
PG 64-22 PG 76-22
Figure 5.22: Impact of Binder Grade on OAC for Tex-204-H Protocol.
The impact of the binder grade on the IDT strengths is shown in Figure 5.24. Unlike the
UCS trends in Figure 5.21, the PG 76-22 specimens seem to be stronger under tension as
compared to the PG 64-22 specimens for almost all materials. The trends for seismic moduli, as
shown in Figure 5.25, are different. The reason for this contradiction is that the modulus is more
indicative of the compressive strength than the tensile strength.
78
137
139
138
139
134
136
140 14
6
136
134
0
50
100
150
200
Beaumont El Paso Houston SH-99
Houston LP-610
Wichita Falls
Den
sity
, pcf
PG 64-22 PG 76-22
Figure 5.23: Impact of Binder Grade on Density for Tex-204-H Protocol.
100
130
114
169
170
135 15
4 167
165
231
0
50
100
150
200
250
300
Beaumont El Paso Houston SH-99
Houston LP-610
Wichita Falls
Indi
rect
Ten
sile
Str
engt
h, p
si
PG 64-22 PG 76-22
Figure 5.24: Impact of Binder Grade on IDT Strength for Tex-204-H Protocol.
79
880 95
0 1064
548
85790
4
842 93
0
550
674
0
500
1000
1500
Beaumont El Paso Houston SH-99
Houston LP-610
Wichita Falls
Seis
mic
Mod
ulus
, ksi
PG 64-22 PG 76-22
Figure 5.25: Impact of Binder Grade on Modulus for Tex-204-H Protocol.
5.4 IMPACT OF AC CONTENT
Each material was tested at its OAC and ±1 of its OAC to observe how the change in
binder content will impact the behavior of the mix. The specimens were cured for 24 hrs and
tested at 77°F. The variations in the UCS with asphalt content for Tex-126-H specimens are
shown in Figure 5.26. Most materials are stronger either at the OAC or 1% less than OAC. A
clear pattern is not obvious because as indicated in Chapter 4 the selection of OAC based on the
density is rather uncertain. The modulus pattern is not that different as shown in Figure 5.27.
The IDT strengths and seismic moduli for the Tex-204-H specimens are shown in Figures
5.28 and 5.29, respectively. In most cases, the change in AC content does not seem to impact the
IDT strength and modulus much.
80
410
591
227
418
342
590
372
584
337
584
387
396
394
380
231
325
331
411
0
100
200
300
400
500
600
700
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Unc
onfin
ed C
ompr
essi
ve S
tren
gth,
psi
(-) 1% OAC % (+) 1 %
Figure 5.26: Impact of Asphalt Content Variation on UCS for Tex-126-H Protocol.
1715
852
1946
2360
1808
2139 2200
1172
2596
2213
201021
51
1229
1994
2461
2001 22
30
2654
0
500
1000
1500
2000
2500
3000
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Seis
mic
Mod
ulus
, ksi
(-) 1% OAC % (+) 1 %
Figure 5.27: Impact of Asphalt Content Variation on Modulus for Tex-126-H Protocol.
81
5.5 IMPACT OF GRADATION
Up to this point, all the specimens were molded using the in place gradation of the
materials as delivered to UTEP. As indicated in Chapter 4, those gradations met the Item 344
gradations. A second gradation that was compatible with Item 292 was developed for each
material by increasing the fine contents in order to quantify the importance of the gradation in a
mix. As reflected in Table 5.3, the modified gradations (values in parenthesis) were obtained by
modifying only the materials finer than No. 40 Sieve so that 15% of the materials passed the
number 200 sieve (while the original gradation contained less than 5%). This experiment was
carried out on four materials.
127
104 111 12
1
109
157
138
185
112
157
109
157
100 11
6
115
115
87
135
0
50
100
150
200
250
300
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Indi
rect
Ten
sile
Str
engt
h, p
si
(-) 1% OAC % (+) 1 %
Figure 5.28: Impact of Asphalt Content Variation on IDT for Tex-204-H Protocol.
82
856
681
669 85
1 946
467
936
1801
844 10
38
994
758819
1105
903 969
949
634
0
500
1000
1500
2000
2500
3000
Beaumont El Paso Houston SH-99
Houston LP-610
Paris WichitaFalls
Seis
mic
Mod
ulus
, ksi
(-) 1% OAC % (+) 1 %
Figure 5.29: Impact of Asphalt Content Variation on Modulus for Tex-204-H Protocol.
5.5.1 Tex-126-H Protocol
Figure 5.30 compares the OACs for the Tex-126-H protocol from the two different
gradations used. The quantities of the asphalt needed to achieve the maximum density for the
Item 292 gradations are higher or close to those from the Item 344 gradations. The changes in
maximum densities are not significant as the fines content increases, but the high fines content
specimens’ densities were found equal or slightly higher than the densities of the original
gradations (see Figure 5.31).
As reflected in Figures 5.32 and 5.33, the addition of fines negatively or positively
impacts the UCS and modulus depending on the original gradations of the materials as retrieved
from the field.
83
Table 5.3: Original and Modified Gradation.
Sieve
Size
Particle
Diameter
(mm)
Beaumont El Paso Houston
SH-99
Houston
LP-610 Paris
Wichita
Falls
2” 58 100 100 100 100 100 100
1.5” 38.1 100 100 100 100 100 100
1" 25.4 83 100 73 91 87 91
7/8" 22.4 76 78 64 88 71 88
3/8" 9.52 56 60 38 63 46 61
#4 4.75 41 45 28 37 30 44
#40 0.425 18 (26)* 23 (26) 19 (22) 20 (26) 14 (22) 16 (25)
#100 0.15 3 (19) 12 (18) 9 (19) 5 (19) 8 (18) 8 (18)
#200 0.075 1 (15) 5 (15) 1 (15) 2 (15) 1 (15) 5 (15) * Numbers in parentheses corresponds to modified gradations that would meet Item 292 requirements.
5.5.2 Tex-204-H Protocol
The OACs for Item 292 and Item 344 gradations for Tex-204-H specimens are compared
in Figure 5.34. The changes in the OACs as a result of changes in gradations seem to be more
pronounced for this protocol as opposed to Tex-126-H. The changes in densities, however, are
again rather small except for the Wichita Falls material (see Figure 5.35).
84
4.7
3.9 4.
1
6.0
4.8
6.0
4.6
6.0
0
2
4
6
8
Beaumont El Paso Houston LP-610
Wichita Falls
Opt
imum
Asp
halt
Con
tent
, %
Avg. 247 Excess Fines
Figure 5.30: Impact of Fines Content Variation on OAC Tex-204-H Protocol.
137 141
138
13313
9 148
140
133
0
50
100
150
200
Beaumont El Paso Houston LP-610
Wichita Falls
Den
sity
, pcf
Avg. 247 Excess Fines
Figure 5.31: Impact of Fines Content Variation on Density Tex-204-H Protocol.
85
174
219
169
301
233
91
195 20
9
0
50
100
150
200
250
300
350
Beaumont El Paso Houston LP-610
Wichita Falls
Unc
onfin
ed C
ompr
essi
ve S
tren
gth,
psi
Avg. 247 Excess Fines
Figure 5.32: Impact of Fines Content Variation on UCS Tex-126-H Protocol.
874
615
1032
1524
879
510
977
883
0
300
600
900
1200
1500
1800
Beaumont El Paso Houston LP-610
Wichita Falls
Seis
mic
Mod
ulus
, ksi
Avg. 247 Excess Fines
Figure 5.33: Impact of Fines Content Variation on Modulus for Tex-126-H Protocol.
86
6.0
3.9
3.8
6.0
5.3
5.1
4.3 4.
6
0
2
4
6
8
Beaumont El Paso Houston LP-610
Wichita Falls
Opt
imum
Asp
halt
Con
tent
, %
Avg. 247 Excess Fines
Figure 5.34: Impact of Fines Content Variation on OAC for Tex – 204 –H.
137
139
139
134
138 142
139
142
0
50
100
150
200
Beaumont El Paso Houston LP-610
Wichita Falls
Den
sity
, pcf
Avg. 247 Excess Fines
Figure 5.35: Impact of Fines Content Variation on Density for Tex – 204 –H.
87
As reflected in Figure 5.36, the IDT strengths from the finer gradations are typically
similar or higher than those from the original, coarser gradations. In Figure 5.37 the moduli
show somewhat different trend. As indicated before, the modulus is more an indication of the
compressive behavior of a mix than tensile behavior.
10
0
130
169
170
206
146 16
5
205
0
50
100
150
200
250
Beaumont El Paso Houston LP-610
Wichita Falls
Indi
rect
Ten
sile
Str
engt
h, p
si
Avg. 247 Excess Fines
Figure 5.36: Impact of Fines Content Variation on IDT for Tex – 204 –H.
88
880 95
0
548
85794
4
1179
900
670
0
300
600
900
1200
1500
1800
Beaumont El Paso Houston LP-610
Wichita Falls
Seis
mic
Mod
ulus
, ksi
Avg. 247 Excess Fines
Figure 5.37: Impact of Fines Content Variation on Modulus for Tex – 204 –H.
89
Chapter Six: Proposed Mix Design Protocol
Along this chapter the proposed mix designed protocol is determined. Many factors were
taken into account to select one of the two proposed methods. Since ATB propose to save
money by reducing the amount of binder, decreasing that amount is primordial in this project, so
the sensitivity of the parameters (density, strength, modulus) to asphalt content is an important
factor to take into account. Other factors were as well considered for availability and ease in
preparation such as: Compaction Device, Specimens Size, Strength Parameter, Curing and
Testing Temperature.
As reflected in the previous chapters, two alternative mix design protocols were
considered. The salient features of these two protocols are compared with those from the
existing Tex-126-E and Tex-204-F in Table 6.1. Based on the evaluation process and
considering many practical aspects, the protocol called Tex-204-H is recommended for the mix
design of the ATB. The preliminary protocol for implementing Tex-204-H is included in
Appendix A.
Table 6.1: Summary of Mix Design Methods Studied.
Mix Design Protocol Tex-126-E Tex-204-F Tex-126-H Tex-204-H
Compactor TGC SGC SGC SGC
Specimen Size, in. 6x8 6x4.5 6x8 6x4.5
OAC based on Asphalt
Density Curve
Volumetric
Properties
Asphalt
Density Curve
Asphalt
Density Curve
Curing Duration and
Temperature
48 hrs
@140°F 48 hrs @77°F
24 hrs
@140°F 24 hrs @ 77°F
Curing/Testing
Temperatures, °F 140 77 77 77
Strength Test UCS IDT UCS IDT
90
6.1 BEST MIX DESIGN SELECTION
A large number of technical and operational factors were considered in this selection.
These factors are discussed below.
• Compaction Device: Because of the availability of the Superpave Gyratory
Compactors (SGC) in all districts and scarcity of the Texas Gyratory Compactors
(TGC), the SGC was selected. This will save funds to refurbish or acquire TGC
devices and minimizes the training of the district technicians.
• Specimen Size: Standard 6 in. x 4.5 in. specimens are recommended because of
ease in preparation, requiring less material for mix design and more uniformity in
specimens.
• Density Calculation: The density is calculated by dividing the weight of the
molded specimen over its volume as done in Item 247, “Flexible Base”.
• Number of Gyrations: Tentatively, a number of gyrations of 75 is proposed. As
shown in Table 6.2, density is not very sensitive to the changes in the asphalt
content for the ATB mixes, which may lead to uncertainty in determining the
OAC. In Table 6.2, sensitivity is defined as the difference between the maximum
and minimum densities at 3%, 4.5% and 6% asphalt contents (used in defining the
asphalt content density relationships) for a given compaction protocol, divided by
the average density of the three measurements. The greater this value is, the
better the asphalt content density curve is defined. The highest sensitivity is
obtained for the number of gyrations of about 80. Since the 75 gyrations are
already included in Tex-204-E, that number was selected. Another angle in
recommending the number of gyrations is the sensitivity of the asphalt content to
strength and modulus. Table 6.2 shows the average sensitivity of these
parameters for Tex-126-H and Tex-204-H protocols for the six different types of
materials used in this study. Again, the highest sensitivities for strength and
modulus are obtained at around 80 gyrations.
91
Table 6.2: MD Curve Sensitivity Results.
* NF – Not Feasible to Compact
• Strength Parameter: Indirect tensile strength tests (IDT) was selected as the
surrogate strength test because of the size of the specimen prepared and sensitivity
of IDT to asphalt content.
• Curing and Testing temperature. Curing temperature of 77oF for 24 hrs and
testing at 77oF are recommended. This will reduce the mix design time and
eliminates the need for additional equipment for high-temperature curing of
specimens (see Chapter 5 for more detail).
• Strength Requirement: A minimum strength of 85 psi as per TxDOT Item 344 is
recommended.
• Moisture Susceptibility: Even though there are some concerns about the moisture
susceptibility of ATB. None of the specimens tested as part of this research
exhibited any signs of stripping or moisture susceptibility. As such, such a
requirement has not been added to the specification.
Density Strength Modulus Gyrations
UCS IDT UCS IDT UCS IDT
60 Gyrations 2% 4% 43% 61% 50% 33%
80 Gyrations 4% 5% 64% 61% 75% 47%
100 Gyrations 2% 3% 48% 45% 55% 22%
TGC 3% NF 68% NF 80% NF*
92
6.2 DETERMINATION OF OAC
In order to determine the OAC the following steps are recommended:
1. Develop IDT-density-asphalt content curves such as the one shown in Figure 6.1.
2. Estimate the maximum density.
3. Convert the density to relative density by dividing the measured densities by the
maximum density as shown in Figure 6.2.
4. Determine the maximum and minimum asphalt contents where the relative
densities are equal or above 97% relative density. An example of this is shown in
Figure 6.3.
5. Determine the maximum and minimum asphalt contents where the IDT strengths
are greater than 85 psi (see Figure 6.3).
6. Determine a range of feasible binder contents as shown in Figure 6.4. The range
of binder content where the relative density is greater than 97% and IDT strength
is greater than 85 psi is recommended. For example, in Figure 6.4, the range is
from 3.3 to 5.4%.
7. Based on the experience of the district and the constructability of the mix, select
the design asphalt content. A value between the traditional OAC and upper limit
range of feasible asphalt contents is recommended (e.g., between 4.4% and 5.4%
in Figure 6.4).
93
131
133
135
137
139
141
143
145
40
55
70
85
100
115
130
3 4 5 6
Density, pcf
Indi
rect
Ten
sile
Str
engt
h, p
si
Asphalt Content, %
Indirect Tensile Strength Density
Figure 6.1: Density/IDT Asphalt Content Combined Curves.
95
96
97
98
99
100
101
40
55
70
85
100
115
130
3 4 5 6
Relative D
ensity, %In
dire
ct T
ensi
le S
tren
gth,
psi
Asphalt Content, %Indirect Tensile Strength Density
Figure 6.2: Relative Density/IDT Asphalt Content Combined Curves.
94
95
96
97
98
99
100
101
40
55
70
85
100
115
130
3 4 5 6
Norm
alized Density, %
Indi
rect
Ten
sile
Str
engt
h, p
si
Asphalt Content, %Indirect Tensile Strength Density
Figure 6.3: Relative Density/IDT Asphalt Content Combined Curves showing Allowable Asphalt Content Zone based on Density.
95
96
97
98
99
100
101
40
55
70
85
100
115
130
3 4 5 6
Norm
alized Density, %
Indi
rect
Ten
sile
Str
engt
h, p
si
Asphalt Content, %Indirect Tensile Strength Density
Figure 6.4: Relative Density/IDT Asphalt Content Combined Curves showing Allowable Asphalt Content Zones based on both Density and IDT.
95
Chapter Seven: Field Performance Monitoring
In this chapter the information obtained from field investigation conducted at several sites
in the El Paso, Houston and Beaumont Districts is presented. Figure 7.1 shows the locations of
the sites. Five sites were monitored before, during and after construction. These sites were
located in Alpine in El Paso District, Beltway 8 and FM 526 (Federal Rd) in Houston District,
and SH 943 and FM 2798 in Beaumont District.
The pavement profile of each site is shown in Figure 7.2. The ATB layers varied in
thickness between 8 in. and 10 in. All sites were covered with surface treatments, expect for
Alpine that the ATB was covered with a concrete slab, and Houston Beltway 8 that the ATB was
not covered at all.
Raw materials were collected from the plants for laboratory tests. During construction,
plant-mixed materials were also collected and taken to the laboratory for index and strength tests.
A Portable Seismic Property Analyzer (PSPA) was used about 24 hrs and 6 months after field
compaction at selected stations. Falling Weight Deflectometer (FWD) tests were performed 6
months after construction also at the selected stations. Cores were extracted from most of the
sites at selected stations. With the material and information gathered at every site, the laboratory
and field properties were compared. The results and conclusions of this activity are presented in
this chapter.
7.1 LABORATORY RESULTS
7.1.1 Raw Materials
The gradation curves of raw materials collected at the quarries are shown in Figure7.3.
Since the same quarry materials were used for both projects in Houston and Beaumont, only one
mix design was carried out for each district. To prepare for mix design from each source, the
entire stock was sieved first to develop a global gradation curve. That gradation curve was used
throughout the study. The acceptable range as per Item 292 Grade 1 is also shown in the figure.
All sites gradation curves met the requirements of Item 292.
96
As shown in Table 7.1, the Alpine materials had a Plasticity Index (PI) of 4 while
Beaumont and Houston materials were non-plastic. The sand equivalency and aggregate
hardness for each material are also presented in Table 7.1. The Houston and Beaumont materials
met the sand equivalency requirements while the Alpine material did not. The Wet Ball Mill test
results are within acceptable range for all sites.
Chapter 6 contains a detailed explanation of the proposed protocol for the determination
of optimum asphalt content (OAC). Three sets of 6 in. x 4.5 in. specimens were prepared with
75 gyrations in a Superpave Gyratory Compactor (SGC) at nominal asphalt contents of 3%, 4.5%
and 6% and tested as per that protocol. Figures 7.4 through 7.6 show the test results for the
Alpine, Beaumont and Houston materials, respectively. Since the variations in the density with
asphalt content are rather small and the IDT strengths in all cases are greater than 85 psi, OAC’s
between 3% and 6% can be used. Based on the current TxDOT practices, an OAC of about
4.5% seems reasonable for all materials. Moduli for each site at different asphalt contents (AC)
are shown in Table 7.2. The greatest moduli were recorded for the Houston material and the
smallest for the Beaumont material. The Alpine material with 3% AC exhibited a very low
modulus because of difficulties in obtaining stable specimens perhaps because of the lack of
fines.
Table 7.1: Sand Equivalency, Wet Ball Mill and Hardness for Materials Used.
Parameter Alpine Beaumont Houston Specification Requirement
Plasticity Index 4 Non Plastic Non Plastic 10 max
Sand Equivalency, % 34 86 63 40 min
Wet Ball Mill, % 27 31 26 50% max
97
a) Alpine Site
b) Beaumont and Houston Sites
Figure 7.1: Locations of Sites Visited used in this Study.
98
02468
1012141618
AlpineBeaumontFM 2798
BeaumontSH 943
HoustonBeltway 8
HoustonFM 526
Thi
ckne
ss, i
n
Seal Coat Concrete ATB Subgrade
Figure 7.2: Pavement Profiles for Sites Visited in this Study.
0102030405060708090
100
0.010.1110100Sieve size, mm
Perc
ent P
assi
ng
Beaumont
Houston
Alpine
#4 #40 #200Gravel Sand Fines
Figure 7.3: Laboratory Gradation Curves for Alpine, Beaumont and Houston Materials.
99
Table 7.2: Variations in Modulus with Asphalt Content for Alpine, Beaumont and Houston Materials.
Sites Alpine Beaumont Houston
AC, % Modulus, ksi
3.0 399 964 1062
4.5 1103 852 1215
6.0 1203 814 1286
7.2 PLANT-MIXED MATERIALS
The binder content of plant-mixed material at each particular site and station was
obtained using an ignition oven. The averages and coefficients of variation (COV) of binder
content are reported in Table 7.3. The Alpine material had the highest average asphalt content
(5.5%). For Houston and Beaumont materials, the average binder contents were close to 4.5%.
The Beaumont SH 943 site exhibited an average AC content of 4.8% with the lowest COV
(3.2%), while the Beaumont FM 2798 site had the lowest AC content with a COV of 5.9%. The
design OAC of each site as provided by the District is also included in Table 7.3 as well. The
asphalt content should not vary by more than 0.5% from the design target. All sites met this
requirement except for the Beaumont SH 943.
Table 7.3: Asphalt Contents of Plant-Mixed Materials.
Binder Content Site
Design, % Average, % COV, %
Alpine 6.0 5.5 4.3
Beaumont FM 2798 4.1 4.3 5.9
Beaumont SH 943 4.1 4.8 3.2
Houston Beltway 8 4.5 4.6 6.2
Houston FM-526 4.5 4.6 9.2
100
45
85
125
165
205
245
285
325
3 4 5 6
Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
s
120
124
128
132
136
140
144
Density, pcf
Indirect Tensile Strength Density
a) Density and IDT vs. AC
45
85
125
165
205
245
285
325
3 4 5 6
Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
s
120
124
128
132
136
140
144
Density, pcf
Indirect Tensile Strength Density
a) Density and IDT vs. AC
45
85
125
165
205
245
285
325
3 4 5 6Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
.
92
93
94
95
96
97
98
99
100
101
Norm
alized Density, %
Indirect Tensile Strength Density
b)Relative Density and IDT vs. AC
45
85
125
165
205
245
285
325
3 4 5 6Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
.
92
93
94
95
96
97
98
99
100
101
Norm
alized Density, %
Indirect Tensile Strength Density
b)Relative Density and IDT vs. AC
Figure 7.1: Density/ Strength vs. Asphalt Content for Alpine Material.
40
55
70
85
100
115
130
3 4 5 6
Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
139
141
143
145D
ensity, pcf
Indirect Tensile Strength Density
a) Density and IDT vs. AC
101
40
55
70
85
100
115
130
3 4 5 6
Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
139
141
143
145D
ensity, pcf
Indirect Tensile Strength Density
a) Density and IDT vs. AC
101
40
55
70
85
100
115
130
3 4 5 6Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
.
92
93
94
95
96
97
98
99
100
101
Norm
alized Density, %
Indirect Tensile Strength Density
b) Relative Density and IDT vs. AC
40
55
70
85
100
115
130
3 4 5 6Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
.
92
93
94
95
96
97
98
99
100
101
Norm
alized Density, %
Indirect Tensile Strength Density
b) Relative Density and IDT vs. AC
Figure 7.2: Density/ Strength vs. Asphalt Content for Houston Material.
40
85
130
175
220
265
310
3 4 5 6
Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
136
138
140
142
Density, pcf
Indirect Tensile Strength Density
a) Density and IDT vs. AC
40
85
130
175
220
265
310
3 4 5 6
Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
136
138
140
142
Density, pcf
Indirect Tensile Strength Density
a) Density and IDT vs. AC
40
85
130
175
220
265
310
3 4 5 6Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
92
93
94
95
96
97
98
99
100
101
Norm
alized Density, %
Indirect Tensile Strength Density
b) Relative Density and IDT vs. AC
40
85
130
175
220
265
310
3 4 5 6Asphalt Content, %
Indi
rect
Ten
sile
Str
engt
h, p
si
92
93
94
95
96
97
98
99
100
101
Norm
alized Density, %
Indirect Tensile Strength Density
b) Relative Density and IDT vs. AC
Figure 7.3: Density/ Strength vs. Asphalt Content for Beaumont Material.
101
Average gradation curves for materials subjected to the ignition oven are presented in
Figure 7.7. Both Alpine gradations (original gradation and the gradation obtained from the plant
mixed materials) met the Item 292 requirements. Even though the Beaumont SH 943 and
Houston FM 526 material gradations obtained from the plant mixed material did not exactly
follow the original design gradation, they still met the Item 292 requirements. The Houston
Beltway 8 and Beaumont FM 2798 plant-mixed materials contained higher sand contents as
compared to their respective original design gradations which mainly consisted of gravel. These
two plant-mixed materials do not meet Item 292 gradation requirements.
Bulk and theoretical maximum specific gravities (Gmb and Gmm, respectively) were
obtained in accordance with Tex-227-F and Tex-201-F for each plant-mixed material. The
average and COV of each parameter, performed on ten specimens for Gmm and six specimens
for Gmb, are shown in Table 7.4. The variability in the specific gravities as judged from COVs
is rather small, and the air voids are all less than 4%.
Three plant-mixed specimens were compacted to nominal dimensions of 6 in. x 4.5 in.
using 75 gyrations of the SGC. These specimens were tested in the same fashion as those
prepared for the proposed mix design (cured at 77°F for 24 hrs). The properties of these plant
mixed specimens are shown in Table 7.5.
Table 7.4: Bulk and Maximum Theoretical Specific Gravities for Plant-Mixed Materials Molded at 75 Gyrations.
Gmm Gmb Site
Average COV, % Average COV, %
Average Air
Voids
Alpine 2.432 0.1 2.427 0.5 1.0%
Beaumont FM 2798 2.441 1.3 2.353 0.3 3.6%
Beaumont SH 943 2.465 0.8 2.377 0.3 3.6%
Houston Beltway 8 2.436 1.3 2.365 0.2 2.9%
Houston FM 526 2.477 1.1 2.404 0.3 3.0%
102
In general, the densities from the five materials are similar with Alpine material being the
densest. The Alpine material exhibited the lowest IDT strength and the highest modulus as
expected from a material with low fines content. Even though, the Beaumont SH 943 and
Houston Beltway 8 specimens exhibited the same density (142 pcf), the Beaumont specimens
showed higher IDT and modulus values as compared to the Houston materials. This may be
explained by the differences in gradations and sand equivalencies.
Table 7.5: Properties of Plant-Mixed Molded using 75 Gyrations of SGC.
Density IDT Seismic Modulus
Site Average,
pcf COV, %
Average,
psi
COV,
%
Average,
ksi
COV,
%
Alpine 147 0.7 150 14.1 1189 1.1
Beaumont FM 2798 140 0.4 214 6.1 886 3.4
Beaumont SH 943 142 0.3 305 2.8 955 3.1
Houston Beltway 8 142 0.4 224 8.7 905 1.1
Houston FM 526 143 0.6 315 4.5 918 4.5
7.3 FIELD CORES
Six cores were extracted from the Alpine site, both Beaumont sites and Houston Beltway
8 site about 24 hrs after compaction. Coring at the Houston FM 526 was not permitted due to the
criticality of the project. The average delivery and compaction temperatures of the mixes at the
approximate locations of the cores are summarized in Table 7.6. The temperatures at delivery
and compaction for the Alpine site were not recorded. According to Item 292 the delivery
temperatures must not exceed 350°F and compaction must be completed before temperature
drops below 175°F. In this case, all the temperatures readings met those requirements.
The densities based on the weight and dimensions and air voids based on the Gmb value
of every core were determined in the laboratory as summarized in Table 7.7. Comparing the
103
results in Tables 7.4 and 7.5 with Table 7.6, the cores’ air voids are significantly higher and their
densities are significantly lower than those of the plant-mixed materials molded by the SGC at
75 gyrations.
Table 7.6: Delivery and Compaction Temperatures of Asphalt Material at the Sites.
Temperature Material
Average, °F COV, %
Delivery 289 3.5 Beaumont FM 2798
Compaction 228 3.0
Delivery 300 5.0 Beaumont SH 943
Compaction 217 7.4
Delivery 273 6.6 Houston Beltway 8
Compaction 209 2.7
Delivery 220 1.4 Houston FM 526
Compaction 202 2.1
Table 7.7: Volumetric Properties of Field Cores.
Density, pcf Gmb
Sites Average COV, % Average COV, %
Air
Voids,
%*
Alpine 145 3.7 2.377 2.3 2.3
Beaumont FM 2798 129 3.9 2.143 4.0 12.2
Beaumont SH 943 124 3.7 2.094 5.2 15.1
Houston Beltway 8 136 4.3 2.288 2.3 6.1 * Gmm’s in Table 7.4 was used to calculate air voids
104
0102030405060708090
100
0.010.1110100
Perc
ent P
assi
ng
Sieve size, mm
Original
Plant-Mixed
Field Cores
#4 #40 #200Gravel Sand Fines
a) Alpine Material
0102030405060708090
100
0.010.1110100
Perc
ent P
assi
ng
Sieve size, mm
Original
Plant-Mixed FM 3513
Field Cores FM 3513
Plant-Mixed FM 2798
Field Cores FM 2798
#4 #40 #200Gravel Sand Fines
b) Beaumont Material
0102030405060708090
100
0.010.1110100
Perc
ent P
assi
ng
Sieve size, mm
Original
Plant-Mixed FM 526
Plant-Mixed Beltway 8
Field Cores Beltway 8
#4 #40 #200Gravel Sand Fines
c) Houston Material
Figure 7.7: Gradation Curves from Plant-Mixed Materials as Delivered.
105
The average modulus and IDT strength of five random cores at each site are shown in
Table 7.8. Alpine and Houston Beltway 8 field cores showed higher IDT strengths and moduli.
Alpine field cores were found to be more uniform since the COVs for IDT and modulus are low.
Table 7.8: Strength and Modulus obtained from Field Cores.
IDT Strength Modulus Site
Average, psi COV, % Average, ksi COV, %
Alpine 145 16.6 891 8.8
Beaumont FM 2798 91 58.0 463 35.9
Beaumont SH 943 86 13.7 380 20.3
Houston Beltway 8 159 29.4 770 18.8
7.4 FIELD RESULTS
The PSPA moduli obtained after 24 hrs and approximately 6 months after construction of
the ATB at each site are shown in Figure 7.8. The PSPA tests could not be performed six
months after construction at Alpine because it was covered with a concrete slab. PSPA moduli
increased in the 6-month period for all sites tested except for Houston Beltway 8 where the ATB
was exposed to the environmental elements and was not trafficked at all. The backcalculated
FWD moduli for all sites are also shown in Figure 7.8. The Alpine FWD results are suspect
because deciphering the moduli of the ATB under a concrete layer is rather difficult.
7.4.1 Comparison of Laboratory and Field Results
The air voids from the plant-mixed materials molded at 75 gyrations using the SGC are
compared with the air voids of the field cores in Figure 7.9. The air voids of the field cores are
considerably higher for all sites as compared to the lab-molded specimens. Similarly, the
densities for the plant-mixed material molded at 75 gyrations are compared to the densities from
106
the field cores in Figure 7.10. Again, substantial differences are observed between the lab and
field densities.
A study was carried out to observe the number of gyrations required to match the field
densities in the lab. It was impossible to simulate the Beaumont SH 943 field density in the lab.
The lowest density that could be obtained by the SGC was 128 pcf by just applying the weight of
the ram. Beaumont FM 2798 and Houston Beltway 8 materials reached their corresponding field
densities in the SGC after 3 and 15 gyrations, respectively. About 66 gyrations were needed to
reach the average field density for the Alpine material.
430385
447516
582584510
463
687
57
350 315
858 830
0
200
400
600
800
1000
Alpine Beaumont FM2798
Beaumont SH943
HoustonBeltway 8
Houston FM526
Des
ign
Mod
ulus
, ksi
PSPA after 24 hours PSPA Design Modulus after 6 Months FWD
Figure 7.8: PSPA and FWD from ATB Sites.
As shown in Figure 7.11, the IDTs of specimens molded at 75 gyrations are naturally
higher at all sites as compared to those from cores extracted from the field. In addition, the IDTs
of the lab specimens compacted to field densities are also marginally to significantly higher than
107
the cores. The same trend is also observed for the seismic moduli of the specimens as shown in
Figure 7.12.
2.6%3.6% 3.6%
2.9%
4.6%
12.2%
15.1%
6.1%
0.0%
4.0%
8.0%
12.0%
16.0%
Alpine Beaumont FM 2798
Beaumont SH 943
Houston Beltway 8
Air
Voi
ds .
Lab Prepared (75 Gyrations) Field Cores
Figure 7.9: Comparison between Air Voids of Laboratory and Field Specimens.
147
140
142
142
145
13614
5
129
128 13
6
129
124
0
30
60
90
120
150
180
Alpine Beaumont FM 2798
Beaumont SH 943
Houston Beltway 8
Den
sity,
pcf
Lab Prepared (75 Gyrations) Field Cores Lab Prepared at Field Density
Figure 7.10: Comparison between Densities of Laboratory and Field Specimens.
108
In Figure 7.12, the moduli obtained from the cores, and PSPA are also included for
comparison. Specimens molded using the 75 gyrations of SGC showed the highest moduli. The
moduli acquired using the V-meter device from the field cores is comparable to the PSPA
modulus 6 months after construction for all. V-meter and PSPA modulus compared in Figure
7.12 do not follow the same pattern because the PSPA modulus shown in this particular figure is
the average of all PSPA modulus readings obtained along the roads.
174
91 86
146
124 14
6 167
214
305
224
145
159
0
50
100
150
200
250
300
350
400
Alpine Beaumont FM 2798
Beaumont SH 943
Houston Beltway 8
Indi
rect
Ten
sile
Stre
ngth
, psi
.
Lab Prepared (75 Gyrations) Field Cores Lab Prepared at Field Density
Figure 7.11: IDT Comparison between Laboratory and Field Specimens.
109
1203
886 95
5
90510
32
859
891
463
380
770
430
385 44
7 51658
4
510
687
704
669
0
500
1000
1500
Alpine Beaumont FM 2798
Beaumont SH 943
Houston Beltway 8
Mod
ulus
, ksi
Lab Prepared (75 Gyrations) Lab Prepared at Field DensityField Cores PSPA after 24 hoursPSPA after 6 Months
Figure 7.12: Modulus Comparison between Laboratory and Field Specimens.
110
Chapter Eight: Conclusions and Recommendations
Based on the overall results of this study, the following conclusions can be outlined:
• The Superpave Gyratory Compactor results were found to be more uniform and
consistent compared to the Texas Gyratory Compactor. Regarding operational
factors, SGC resulted to be easier and more efficient compared to the TGC.
• When determining the OAC of an ATB material following the method explained
in this report using the SGC resulted to be lower than the OAC obtained from the
TGC.
• The asphalt content variations caused no significant changes in density of molded
specimens using the SGC.
• Specimens molded at 80 gyrations were found to have the highest sensitivity to
asphalt content compared to the specimens molded at 60 and 100 gyrations. Since
the 75 gyrations are already included in Tex-204-E; 75 was selected as the
number of gyrations.
• Superpave Gyratory Compactor specimens usually exhibit similar or higher UCS
as compared to those prepared with Texas Gyratory Compactor.
• Superpave Gyratory Compactor specimens usually exhibit similar or higher IDT
as compared to those prepared with Texas Gyratory Compactor
• Specimen Standard Size 6 in. in diameter and 4.5 in. in height were selected to be
the more efficient because of the ease in preparation; they require less material for
mix design and resulted to have a higher uniformity.
• Indirect Tensile Strength was found to be more sensitive to the asphalt content, so
it was selected to be as a parameter to select the OAC of a mix design. The OAC
was selected to be any value in a standard Asphalt-Content Curve which meets
the following two requisites:
o 97% of the relative density
111
o 85 psi of IDT as per TxDOT Item 344 is recommended
Curing temperature of 77°F during 24 hours after compaction is recommended. Testing
temperature of 77°F is also recommended.
112
References
[AASHTO, Designation T269, “Percent Air Voids in Compacted Dense and Open Asphalt Mixtures” Standard Specifications for Transportation Materials and Methods of Sampling and Testing, 25th Edition, AASHTO, Washington, DC, 2005, CD-ROM.]
[Aguiar-Moya, J. P., Prozzi, J. A., Tahmoressi, M. “Optimum Number of Superpave Gyrations Based on Project Requirements”. Transportation Research Board, Washington D.C. 2007]
[Azari, H., Lutz, R., and Spellberg, P., “Precision Estimates of Selected Volumetric Properties of HMA Using Absorptive Aggregate,” Submitted for Approval by the NCHRP 9-26 Panel. 2006.]
[Azari, H., Lutz, R., And Spellberg, P., “Precision Estimates for AASHTO Test Method T 269 Determined Using AMRL Proficiency Sample Data,” NCHRP Web Document 114, 2007.]
[Button, J. W., Chowdhury, A., and Bhasin A. (2006) “Transitioning from the Texas Gyratory Compactor to the Superpave Gyratory Compactor” 86th Annual Meeting of the Transportation Research Board, Washington, D.C., January 2006.]
[Button, J. W., Little, D.N., Jagadam, V., and Pendleton, O.J. “correlation of Selected Laboratory Compaction Methods with Field Compaction,” Transportation Research Record 1454, Transportation Research Board, Washington D.C., 1994, pp. 193-201.]
[Chakroborty, P., and Das, A. Principles of Transportation Engineering PHI Learning Pvt. Ltd., pp 360. 2004.]
[Cedergren, H.R. Drainage of Highway and Airfield Pavements, John Wiley & Sons, New York. 1977.]
[Collins, R., Watson, D., Johnson, A., and Wu, Y. Effect of Aggregate Degradation of Specimens Compacted by Superpave Gyratory Compactor. In Transportation Research Record: Journal of the Transportation Research Board, No. 1590, TRB, National Research Council, Washington D.C., 1997, pp.1-9.]
[Dykman, M., Ramanujam, J.M., and Nataatmadja, A. (2003) “Performance of Bitumen Treated Bases” 82nd Annual Meeting of the Transportation Research Board, Washington, D.C., January 2003]
[Harman, T., Bukowski, J.R., Mountier, F., Huber, G., and McGennis, R. “The History and Future Challenges Of Gyratory Compaction 1939 to 2001,” Transportation Research Board, National Research council, Washington, D.C., 2002]
[Holsinger, R.E., Fisher, A., and Spellberg, P.A., “Precision Estimates for AASHTO Test Method T308 and the Test Methods for Performance-Graded Asphalt Binder in AASHTO Specification M320,” NCHRP Web Document 71, 2005]
[Marks, V.J. and Huisman, C.L. Reducing the Adverse Effects of Transverse Cracking, Transportation Research Record, 1034, pp 80-86. 1985]
113
[McDowell, C., and Smith, A.W. “Design, Control, and Interpretation of Tests for Bituminous Hot Mix Black Base Mixtures”. Materials and Test Division, Soils Section Texas Highway Department, 1969]
[Mokwa, R., Cuelho, E., Browne, M. “Laboratory Testing of Soils Using the Superpave yratory Compactor” 87th Annual Meeting of the Transportation Research Board, Washington, D.C., January 2008]
[Peterson, R. L., Mahboub, K. C., Anderson, R. M., Masad, E., Tashman, T. Comparing Superpave Gyratory Compactor Data to Field Cores. Journal of Materials in Civil Engineering ASCE. 2004]
[Saeed A., Hall, J.W., and Barker W., “Performance-Related Tests of Aggregates for Use in Unbound Pavement Layers,” NCHRP Report 453, national Academy of Engineering, Washington, DC. 2001]
[Sebesta, S., Guthrie, W.S., and Harris, J.P. “Gyratory Compaction of Soils for Laboratory Swell Tests,” Proceedings; 83rd Annual Meeting of the Transportation Research Board, Washington, D.C., January 2004]
[Spellberg, P.A., and Savage, D.A., “An Investigation of the Cause of Variation in HMA Bulk Specific Gravity Test Results Using Non-Absorptive Aggregates,” NCHRP Web Document 66, 2004]
[Spellberg, P.A., Savage, D.A., and Pielert, J.H., “Precision Estimates of Selected Volumetric Properties of HMA Using Non-Absorptive Aggregate” NCHRP Web Document 54, 2003]
[Ullman & Nolan Consulting Engineers. Report on Design Characteristic for Logan City Council. Reference 1419U\PJD\8.91.1991.]
[Vavrik, W. R., and S. H. Carpenter, “Calculating Air Voids at Specified Numbers of Gyrations in Superpave Gyratory Compactor,” Transportation Research Record 1630, 1998.]
[Von Quintus, H., Scherocman, J.A., Hughes, C.W., and Kennedy, T.W. “Asphalt-Aggregate Mixture Analysis System –AAMAS,” NCHRP Report 338, national cooperative Highway Research Program, Transportation Research Board, Washington, D.C., 1991]
[Wong, W.G., Yang, Q., and Wang, K.C.P. “Performance-Based Mixture Design of Asphalt-Treated Base,” Journal of the Institution of Engineers, Vol. 44 Issue 2, Singapore. 2004]
114
Curriculum Vita
Hector A. Hernandez was born on November 30, 1985 in Chihuahua, Chihuahua,
Mexico, the second son of Humberto A. Hernandez and Guillermina Sandoval Ortiz. He
graduated from Escuela Federal por Cooperacion in the spring of 2004 and entered University of
Texas at El Paso (UTEP) in the fall of the same year. He received his bachelor’s degree in Civil
Engineering from University of Texas at El Paso in 2009. While pursuing his bachelor’s degree
he joined many honors societies such as: National Honors Society, Alpha Chi, and Chi Epsilon.
He is an Engineer in Training by the Texas Board of Professional Engineers (2009). In the
spring of 2010, he entered the graduate program of the Civil Engineering Department at UTEP.
He received the SemMaterial Scholarship and the Eisenhower Fellowship in the spring of 2009
and 2010, respectively for his academic achievements. He graduated with a Grade Point
Average of 3.7 and 3.8 during his undergraduate and graduate studies, respectively. Upon
graduation, he was prized with the Andrew Jones Award for his performance during his
undergraduate studies.
During his undergraduate studies he joined the Center for Transportation Infrastructure
Systems (CTIS) at UTEP as an Undergraduate Research Assistant. After completion of his
undergraduate studies, he became Graduate Research Assistant. He started working on a solar
pavement markers project and then continued working in project named: “Development of a
New Mix Design Method and Specification Requirements for Asphalt Treated Base”, funded by
the Texas Department of Transportation. He worked at CTIS for approximately 3 years. He is
currently a Geotechnical Project Engineer at the Construction Quality Control (CQC) company
in El Paso, Texas.
Permanent address: Avenida Tercera Oriente #703
Delicias, Chihuahua, Mexico, 33000
This thesis was typed by Hector A. Hernandez.